1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.404 2006/10/09 14:36:25 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TF/01 In the add_headers option to the mail command in an Exim filter,
10 there was a bug that Exim would claim a syntax error in any
11 header after the first one which had an odd number of characters
14 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
15 callout verification, Exim cached a "reject" for the entire domain. This
16 is correct for most verifications, but it is not correct for a recipient
17 verification with use_sender or use_postmaster set, because in that case
18 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
19 case of MAIL FROM:<> rejection from other early rejections (e.g.
20 rejection of HELO). When verifying a recipient using a non-null MAIL
21 address, the cache is ignored if it shows MAIL FROM:<> rejection.
22 Whatever the result of the callout, the value of the domain cache is
23 left unchanged (for any other kind of callout, getting as far as trying
24 RCPT means that the domain itself is ok).
26 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
29 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
30 manifest itself as EPIPE rather than ECONNECT. When tidying away a
31 session, the daemon ignores ECONNECT errors and logs others; it now
32 ignores EPIPE as well.
34 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
35 (quoted-printable decoding).
37 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
38 later the small subsequent patch to fix an introduced bug.
40 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
42 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
44 PH/08 An error is now given if message_size_limit is specified negative.
46 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
47 to be given (somewhat) arbitrary names.
49 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
52 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
53 miscellaneous code fixes
55 PH/10 Added the log_reject_target ACL modifier to specify where to log
58 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
59 hostname. This is wrong, because it relates to the incoming message (and
60 probably the interface on which it is arriving) and not to the outgoing
61 callout (which could be using a different interface). This has been
62 changed to use the value of the helo_data option from the smtp transport
63 instead - this is what is used when a message is actually being sent. If
64 there is no remote transport (possible with a router that sets up host
65 addresses), $smtp_active_hostname is used.
67 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
68 tweaks were necessary in order to get it to work:
69 (a) The code assumed that strncpy() returns a negative number on buffer
70 overflow, which isn't the case. Replaced with Exim's string_format()
72 (b) There were several signed/unsigned issues. I just did the minimum
73 hacking in of casts. There is scope for a larger refactoring.
74 (c) The code used strcasecmp() which is not a standard C function.
75 Replaced with Exim's strcmpic() function.
76 (d) The code set only $1; it now sets $auth1 as well.
77 (e) A simple test gave the error "authentication client didn't specify
78 service in request". It would seem that Dovecot has changed its
79 interface. Fortunately there's a specification; I followed it and
80 changed what the client sends and it appears to be working now.
82 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
85 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
86 address A is aliased to B and C, where B exists and C does not. Without
87 -v the output is "A verified" because verification stops after a
88 successful redirection if more than one address is generated. However,
89 with -v the child addresses are also verified. Exim was outputting "A
90 failed to verify" and then showing the successful verification for C,
91 with its parentage. It now outputs "B failed to verify", showing B's
92 parentage before showing the successful verification of C.
94 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
95 look up a TXT record in a specific list after matching in a combined
98 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
99 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
100 they consult the DNS. I had assumed they would set it the way they
101 wanted; and indeed my experiments on Linux seem to show that in some
102 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
103 To be on the safe side, however, I have now made the interface to
104 host_find_byname() similar to host_find_bydns(), with an argument
105 containing the DNS resolver options. The host_find_byname() function now
106 sets these options at its start, just as host_find_bydns() does. The smtp
107 transport options dns_qualify_single and dns_search_parents are passed to
108 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
109 of host_find_byname() use the default settings of RES_DEFNAMES
110 (qualify_single) but not RES_DNSRCH (search_parents).
116 SC/01 Use a glob alias rather than an array ref in eximstats generated
117 parser. This improves both readability and performance.
119 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
120 Don't display local sender or destination tables in eximstats unless
121 there is data to show.
122 Added average volumes into the eximstats top table text output.
124 SC/03 Collect data on the number of addresses (recipients) as well
125 as the number of messages in eximstats.
127 TF/01 Correct an error in the documentation for the redirect router. Exim
128 does (usually) call initgroups() when daemonizing.
130 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
131 with consistent privilege compared to when running as a daemon.
133 TF/03 Note in the spec that $authenticated_id is not set for local
134 submissions from trusted users.
136 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
137 Thanks to Dean Brooks <dean@iglou.com> for the patch.
139 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
140 by adding some example configuration directives to the default
141 configuration file. A little bit of work is required to uncomment the
142 directives and define how usernames and passwords are checked, but
143 there is now a framework to start from.
145 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
146 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
147 without this. I don't know how relevant this is to other LDAP libraries.
149 PH/02 Add the verb name to the "unknown ACL verb" error.
151 PH/03 Magnus Holmgren's patch for filter_prepend_home.
153 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
155 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
156 directory not expanded when it should be if an expanded home directory
157 was set for the address (which is overridden by the transport).
159 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
162 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
163 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
164 because it is too late at that time, and has no effect.
166 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
167 security issue with \' (bugzilla #107). I could not use the
168 PQescapeStringConn() function, because it needs a PGconn value as one of
171 PH/08 When testing addresses using -bt, indicate those final addresses that
172 are duplicates that would not cause an additional delivery. At least one
173 person was confused, thinking that -bt output corresponded to deliveries.
174 (Suppressing duplicates isn't a good idea as you lose the information
175 about possibly different redirections that led to the duplicates.)
177 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
178 systems where poll() doesn't work, in particular OS X.
180 PH/10 Added more information to debugging output for retry time not reached.
182 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
183 operations in malware.c.
185 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
188 PH/13 If write_rejectlog was set false when logging was sent to syslog with
189 syslog_duplication set false, log lines that would normally be written
190 both the the main log and to the reject log were not written to syslog at
193 PH/14 In the default configuration, change the use of "message" in ACL warn
194 statements to "add_header".
196 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
197 not followed by a command (e.g. "seen endif").
199 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
200 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
203 PH/17 Added extra conditions to the default value of delay_warning_condition
207 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
208 { match{$h_precedence:}{(?i)bulk|list|junk} } \
209 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
212 The Auto-Submitted: and various List- headers are standardised, whereas I
213 don't think Precedence: ever was.
215 PH/18 Refactored debugging code in route_finduser() to show more information,
216 in particular, the error code if getpwnam() issues one.
218 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
219 This is apparently needed in addition to the PH/07 change above to avoid
220 any possible encoding problems.
222 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
223 but not after initializing Perl.
225 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
226 output them only if debugging. By default they are written stderr,
227 apparently, which is not desirable.
229 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
232 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
235 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
237 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
238 authenticated or an ident call has been made. Suppress the default
239 values for $authenticated_id and $authenticated_sender (but permit -oMai
240 and -oMas) when testing with -bh.
242 PH/24 Re-jigged the order of the tests in the default configuration so that the
243 tests for valid domains and recipients precede the DNS black list and CSA
244 tests, on the grounds that those ones are more expensive.
246 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
247 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
248 This bug exists in every version of Exim that I still have, right back to
251 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
252 However, an attempt to turn on case-sensitivity in a regex key by
253 including (?-i) didn't work because the subject string was already
254 lowercased, and the effects were non-intuitive. It turns out that a
255 one-line patch can be used to allow (?-i) to work as expected.
261 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
262 other effects) broke the use of negated acl sub-conditions.
264 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
267 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
268 "Deny" causes Exim to reject the incoming connection with a 554 error.
269 Unfortunately, if there is a major crisis, such as a disk failure,
270 tcp-wrappers gives "deny", whereas what one would like would be some
271 kind of temporary error. A kludge has been added to help with this.
272 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
273 554 error is used if errno is still zero or contains ENOENT (which occurs
274 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
277 PH/03 Add -lutil to the default FreeBSD LIBS setting.
279 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
280 errors. Otherwise a message that provokes a temporary error (when other
281 messages do not) can cause a whole host to time out.
283 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
284 addresses were routed directly to files or pipes from a redirect router.
285 File deliveries just didn't batch; pipe deliveries might have suffered
288 PH/06 A failure to get a lock for a hints database would erroneously always say
289 "Failed to get write lock", even when it was really a read lock.
291 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
292 of 0600. This has been changed to use the value of the lockfile_mode
293 option (which defaults to 0600).
295 PH/08 Applied small patch from the Sieve maintainer.
297 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
298 folder from quota calculations, a direct delivery into this folder messed
299 up the contents of the maildirsize file. This was because the regex was
300 used only to exclude .Trash (or whatever) when the size of the mailbox
301 was calculated. There was no check that a delivery was happening into an
302 excluded directory. This bug has been fixed by ignoring all quota
303 processing for deliveries into excluded directories.
305 PH/10 Added the maildirfolder_create_regex option to appendfile.
311 PH/01 The code for finding all the local interface addresses on a FreeBSD
312 system running IPv6 was broken. This may well have applied to all BSD
313 systems, as well as to others that have similar system calls. The broken
314 code found IPv4 interfaces correctly, but gave incorrect values for the
315 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
316 that it would not match correctly against @[] and not recognize the IPv6
319 PH/02 The ipliteral router was not recognizing addresses of the form user@
320 [ipv6:....] because it didn't know about the "ipv6:" prefix.
322 PH/03 Added disable_ipv6.
324 PH/04 Changed $reply_address to use the raw form of the headers instead of the
325 decoded form, because it is most often used to construct To: headers
326 lines in autoreplies, and the decoded form may well be syntactically
327 invalid. However, $reply_address has leading white space removed, and all
328 newlines turned into spaces so that the autoreply transport does not
331 PH/05 If group was specified without a user on a router, and no group or user
332 was specified on a transport, the group from the router was ignored.
334 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
335 for visible compile-time settings that can be used to change these
336 numbers, for those that want even more. Backwards compatibility with old
337 spool files has been maintained. However, going back to a previous Exim
338 release will lost any variables that are in spool files.
340 PH/07 Two small changes when running in the test harness: increase delay when
341 passing a TCP/IP connection to a new process, in case the original
342 process has to generate a bounce, and remove special handling of
343 127.0.0.2 (sic), which is no longer necessary.
345 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
346 be the same on different OS.
348 PH/09 Moved a debug statement in filter processing to avoid a race problem when
351 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
352 whether --show-vars was specified or not
354 JJ/02 exipick: Added support for new ACL variable spool format introduced
357 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
358 syntactically invalid From: or Reply-to: line, and a filter used this to
359 generate an autoreply, and therefore failed to obtain an address for the
360 autoreply, Exim could try to deliver to a non-existent relative file
361 name, causing unrelated and misleading errors. What now happens is that
362 it logs this as a hard delivery error, but does not attempt to create a
365 PH/11 The exinext utility has a -C option for testing purposes, but although
366 the given file was scanned by exinext itself; it wasn't being passed on
367 when Exim was called.
369 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
370 an end-of-file indication when reading a command response.
372 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
373 compiled. In many other places in Exim, IPv6 addresses are always
374 recognized, so I have changed this. It also means that IPv4 domain
375 literals of the form [IPV4:n.n.n.n] are now always recognized.
377 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
378 used if the router is not running as root, for example, when verifying at
379 ACL time, or when using -bh. The debugging output from this situation was
380 non-existent - all you got was a failure to exec. I have made two
383 (a) Failures to set uid/gid, the current directory, or a process leader
384 in a subprocess such as that created by queryprogram now generate
385 suitable debugging ouput when -d is set.
387 (b) The queryprogram router detects when it is not running as root,
388 outputs suitable debugging information if -d is set, and then runs
389 the subprocess without attempting to change uid/gid.
391 PH/15 Minor change to Makefile for building test_host (undocumented testing
394 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
395 additional section of a DNS packet that returns MX or SRV records.
396 Instead, it always explicitly searches for A/AAAA records. This avoids
397 major problems that occur when a DNS server includes only records of one
398 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
399 fixed another bug: if SRV records were looked up and the corresponding
400 address records were *not* found in the additional section, the port
401 values from the SRV records were lost.
403 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
404 using the correct key (the original address) when searching the retry
405 rules in order to find which one to use for generating the retry hint.
407 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
408 adding the default one. Similarly, if it contains a Reply-To: header, the
409 errors_reply_to option, if set, is not used.
411 PH/19 When calculating a retry time, Exim used to measure the "time since
412 failure" by looking at the "first failed" field in the retry record. Now
413 it does not use this if it is later than than the arrival time of the
414 message. Instead it uses the arrival time. This makes for better
415 behaviour in cases where some deliveries succeed, thus re-setting the
416 "first failed" field. An example is a quota failure for a huge message
417 when small messages continue to be delivered. Without this change, the
418 "time since failure" will always be short, possible causing more frequent
419 delivery attempts for the huge message than are intended.
420 [Note: This change was subsequently modified - see PH/04 for 4.62.]
422 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
423 $1, $2, $3) because the numerical variables can be reset during some
424 expansion items (e.g. "match"), thereby losing the authentication data.
426 PH/21 Make -bV show the size of off_t variables so that the test suite can
427 decide whether to run tests for quotas > 2G.
429 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
430 mailbox_size, and mailbox_filecount in the appendfile transport. If a
431 filecount value is greater than 2G or if a quota value is greater than 2G
432 on a system where the size of off_t is not greater than 4, a panic error
435 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
436 never match. The debug and -bh output now contains an explicit error
437 message indicating a malformed IPv4 address or mask.
439 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
440 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
443 PH/25 Do not write to syslog when running in the test harness. The only
444 occasion when this arises is a failure to open the main or panic logs
445 (for which there is an explicit test).
447 PH/26 Added the /no_tell option to "control=freeze".
449 PH/27 If a host name lookup failed very early in a connection, for example, if
450 the IP address matched host_lookup and the reverse lookup yielded a name
451 that did not have a forward lookup, an error message of the form "no IP
452 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
453 could be logged. Now it outputs the IP address instead of "NULL".
455 PH/28 An enabling patch from MH: add new function child_open_exim2() which
456 allows the sender and the authenticated sender to be set when
457 submitting a message from within Exim. Since child_open_exim() is
458 documented for local_scan(), the new function should be too.
460 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
461 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
462 results in an empty string is now treated as unset.
464 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
466 PH/31 Added sender_verify_fail logging option.
468 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
469 needed by RFC 822 but not by RFC 2822 was commented out. I have now
470 tidied the source and removed it altogether.
472 PH/33 When a queue run was abandoned because the load average was too high, a
473 log line was always written; now it is written only if the queue_run log
474 selector is set. In addition, the log line for abandonment now contains
475 information about the queue run such as the pid. This is always present
476 in "start" and "stop" lines but was omitted from the "abandon" line.
478 PH/34 Omit spaces between a header name and the colon in the error message that
479 is given when verify = headers_syntax fails (if there are lots of them,
480 the message gets confusing).
482 PH/35 Change the default for dns_check_names_pattern to allow slashes within
483 names, as there are now some PTR records that contain slashes. This check
484 is only to protect against broken name servers that fall over on strange
485 characters, so the fact that it applies to all lookups doesn't matter.
487 PH/36 Now that the new test suite is complete, we can remove some of the
488 special code in Exim that was needed for the old test suite. For example,
489 sorting DNS records because real resolvers return them in an arbitrary
490 order. The new test suite's fake resolver always returns records in the
493 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
494 bounces) except when queue_only is set, to avoid logging races between
495 the different processes.
497 PH/38 Panic-die if .include specifies a non-absolute path.
499 PH/39 A tweak to the "H" retry rule from its user.
501 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
502 a label. They prevented compilation on older perls.
504 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
505 a warning to be raised on newish perls.
507 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
508 on queue. Changes to match documented behaviour of showing count of
509 messages matching specified criteria.
511 PH/40 Changed the default ident timeout from 30s to 5s.
513 PH/41 Added support for the use of login_cap features, on those BSD systems
514 that have them, for controlling the resources used by pipe deliveries.
516 PH/42 The content-scanning code uses fopen() to create files in which to put
517 message data. Previously it was not paying any attention to the mode of
518 the files. Exim runs with umask(0) because the rest of the code creates
519 files with open(), and sets the required mode explicitly. Thus, these
520 files were ending up world-writeable. This was not a big issue, because,
521 being within the spool directory, they were not world-accessible. I have
522 created a function called modefopen, which takes an additional mode
523 argument. It sets umask(777), creates the file, chmods it to the required
524 mode, then resets the umask. All the relevant calls to fopen() in the
525 content scanning code have been changed to use this function.
527 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
528 to 24 hours. This avoids potential overflow problems when processing G
529 and H retry rules. I suspect nobody ever tinkers with this value.
531 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
533 PH/45 When the plaintext authenticator is running as a client, the server's
534 challenges are checked to ensure they are valid base64 strings. By
535 default, the authentication attempt is cancelled if an invalid string is
536 received. Setting client_ignore_invalid_base64 true ignores these errors.
537 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
538 they are received. Thus, the responses can be made to depend on the
539 challenges. If an invalid string is ignored, an empty string is placed in
542 PH/46 Messages that are created by the autoreply transport now contains a
543 References: header, in accordance with RFCs 2822 and 3834.
545 PH/47 Added authenticated_sender_force to the smtp transport.
547 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
549 PH/49 Installed latest patch from the Sieve maintainer.
551 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
552 was also set, the appendfile transport was unnecessarily scanning a
553 directory of message files (e.g. for maildir delivery) to find the count
554 of files (along with the size), even though it did not need this
555 information. It now does the scan only if it needs to find either the
556 size of the count of files.
558 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
560 PH/52 Two bugs concerned with error handling when the smtp transport is
563 (i) Exim was not creating retry information for temporary errors given
564 for individual recipients after the DATA command when the smtp transport
565 was used in LMTP mode. This meant that they could be retried too
566 frequently, and not timed out correctly.
568 (ii) Exim was setting the flag that allows error details to be returned
569 for LMTP errors on RCPT commands, but not for LMTP errors for individual
570 recipients that were returned after the DATA command.
572 PH/53 This is related to PH/52, but is more general: for any failing address,
573 when detailed error information was permitted to be returned to the
574 sender, but the error was temporary, then after the final timeout, only
575 "retry timeout exceeded" was returned. Now it returns the full error as
576 well as "retry timeout exceeded".
578 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
579 do this, and (what is worse) MTAs that accept it.
581 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
582 will now be deprecated.
584 PH/56 New os.c-cygwin from the Cygwin maintainer.
586 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
587 formats (previously only available in exim formats via -bpr, -bpru,
588 and -bpra. Now also available in native and exiqgrep formats)
590 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
591 with very large, slow to parse queues
593 JJ/08 exipick: added ! as generic prefix to negate any criteria format
595 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
597 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
598 responses to authentication challenges, though it was showing the
599 challenges; (ii) I've removed the CR characters from the debug output for
602 PH/58 Allow for the insertion of a newline as well as a space when a string
603 is turned into more than one encoded-word during RFC 2047 encoding. The
604 Sieve code now uses this.
606 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
607 data_4xx, lost_connection, tls_required.
609 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
610 message was being sent as an SMTP response.
612 PH/61 Add -l and -k options to exicyclog.
614 PH/62 When verifying, if an address was redirected to one new address, so that
615 verification continued, and the new address failed or deferred after
616 having set something in $address_data, the value of $address_data was not
617 passed back to the ACL. This was different to the case when no
618 redirection occurred. The value is now passed back in both cases.
620 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
621 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
622 use login_cap.h, so on its own it isn't the distinguishing feature. The
623 new name refers directly to the setclassresources() function.
625 PH/65 Added configuration files for NetBSD3.
627 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
629 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
630 is preferred over IPv4.
632 PH/68 The bounce_return_message and bounce_return_body options were not being
633 honoured for bounces generated during the reception of non-SMTP messages.
634 In particular, this applied to messages rejected by the ACL. This bug has
635 been fixed. However, if bounce_return_message is true and bounce_return_
636 body is false, the headers that are returned for a non-SMTP message
637 include only those that have been read before the error was detected.
638 (In the case of an ACL rejection, they have all been read.)
640 PH/69 The HTML version of the specification is now built in a directory called
641 spec_html instead of spec.html, because the latter looks like a path with
642 a MIME-type, and this confuses some software.
644 PH/70 Catch two compiler warnings in sieve.c.
646 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
647 function verify_get_ident() calls ip_connect() to connect a socket, but
648 if the "connect()" function timed out, ip_connect() used to close the
649 socket. However, verify_get_ident() also closes the socket later, and in
650 between Exim writes to the log, which may get opened at this point. When
651 the socket was closed in ip_connect(), the log could get the same file
652 descriptor number as the socket. This naturally causes chaos. The fix is
653 not to close the socket in ip_connect(); the socket should be closed by
654 the function that creates it. There was only one place in the code where
655 this was missing, in the iplookup router, which I don't think anybody now
656 uses, but I've fixed it anyway.
658 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
659 well as to direct DNS lookups. Otherwise the handling of names in host
660 lists is inconsistent and therefore confusing.
666 PH/01 Two changes to the default runtime configuration:
668 (1) Move the checks for relay_from_hosts and authenticated clients from
669 after to before the (commented out) DNS black list checks.
671 (2) Add control=submission to the relay_from_hosts and authenticated
672 clients checks, on the grounds that messages accepted by these
673 statements are most likely to be submissions.
675 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
677 (1) Generate an error if the third argument for the ${prvs expansion is
680 (2) Treat a missing third argument of ${prvscheck as if it were an empty
683 (3) Reset the variables that are obtained from the first argument of
684 ${prvscheck and used in the second argument before leaving the code,
685 because their memory is reclaimed, so using them afterwards may do
688 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
689 one (it's much easier than Tom thought :-).
691 (5) Because of (4), we can now allow for the use of $prvscheck_result
692 inside the third argument.
694 PH/03 For some reason, the default setting of PATH when running a command from
695 a pipe transport was just "/usr/bin". I have changed it to
698 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
699 anything to be listed in the output from -bV.
701 PH/05 When a filter generated an autoreply, the entire To: header line was
702 quoted in the delivery log line, like this:
704 => >A.N.Other <ano@some.domain> <original@ddress> ...
706 This has been changed so that it extracts the operative address. There
707 may be more than one such address. If so, they are comma-separated, like
710 => >ano@some.domain,ona@other.domain <original@ddress> ...
712 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
713 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
714 not being looked up in the rDNS to get a host name, Exim was showing the
715 IP address twice in Received: lines, even though the IP addresses were
716 identical. For example:
718 Received: from [1.2.3.4] (helo=[1.2.3.4])
720 However, if the real host name was known, it was omitting the HELO data
721 if it matched the actual IP address. This has been tidied up so that it
722 doesn't show the same IP address twice.
724 PH/07 When both +timestamp and +memory debugging was on, the value given by
725 $tod_xxx expansions could be wrong, because the tod_stamp() function was
726 called by the debug printing, thereby overwriting the timestamp buffer.
727 Debugging no longer uses the tod_stamp() function when +timestamp is set.
729 PH/08 When the original message was included in an autoreply transport, it
730 always said "this is a copy of the message, including all the headers",
731 even if body_only or headers_only was set. It now gives an appropriate
734 PH/09 Applied a patch from the Sieve maintainer which:
736 o fixes some comments
737 o adds the (disabled) notify extension core
738 o adds some debug output for the result of if/elsif tests
739 o points to the current vacation draft in the documentation
740 and documents the missing references header update
744 o fixes a bug in processing the envelope test (when testing
745 multiple envelope elements, the last element determinted the
748 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
749 Electronic Mail") by including:
751 Auto-submitted: auto-generated
753 in the messages that it generates (bounce messages and others, such as
754 warnings). In the case of bounce messages for non-SMTP mesages, there was
755 also a typo: it was using "Auto_submitted" (underscore instead of
756 hyphen). Since every message generated by Exim is necessarily in response
757 to another message, thes have all been changed to:
759 Auto-Submitted: auto-replied
761 in accordance with these statements in the RFC:
763 The auto-replied keyword:
765 - SHOULD be used on messages sent in direct response to another
766 message by an automatic process,
768 - MUST NOT be used on manually-generated messages,
770 - MAY be used on Delivery Status Notifications (DSNs) and Message
771 Disposition Notifications (MDNs),
773 - MUST NOT be used on messages generated by automatic or periodic
774 processes, except for messages which are automatic responses to
777 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
778 to the default Received: header definition.
780 PH/12 Added log selector acl_warn_skipped (default on).
782 PH/13 After a successful wildlsearch lookup, discard the values of numeric
783 variables because (a) they are in the wrong storage pool and (b) even if
784 they were copied, it wouldn't work properly because of the caching.
786 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
787 checking when decoding. Apparently there are clients that generate
788 overlong encoded strings. Why am I not surprised?
790 PH/15 If the first argument of "${if match_address" was not empty, but did not
791 contain an "@" character, Exim crashed. Now it writes a panic log message
792 and treats the condition as false.
794 PH/16 In autoreply, treat an empty string for "once" the same as unset.
796 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
797 extension "envelope-auth". The code is finished and in agreement with
798 other implementations, but there is no documentation so far and in fact,
799 nobody wrote the draft yet. This extension is currently #undef'ed, thus
800 not changing the active code.
802 Print executed "if" and "elsif" statements when debugging is used. This
803 helps a great deal to understand what a filter does.
805 Document more things not specified clearly in RFC3028. I had all this
806 sorted out, when out of a sudden new issues came to my mind. Oops."
808 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
811 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
812 canonical form (as documented). However, after a host name lookup from
813 the IP address, check_host() was doing a simple string comparison with
814 addresses acquired from the DNS when checking that the found name did
815 have the original IP as one of its addresses. Since any found IPv6
816 addresses are likely to be in abbreviated form, the comparison could
817 fail. Luckily, there already exists a function for doing the comparison
818 by converting both addresses to binary, so now that is used instead of
821 PH/20 There was another similar case to PH/19, when a complete host name was
822 given in a host list; looking up its IP address could give an abbreviated
823 form, whereas the current host's name might or might not be abbreviated.
824 The same fix has been applied.
830 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
831 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
834 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
835 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
837 PH/03 Typo: missing ".o" in src/pcre/Makefile.
839 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
840 header line, restrict the check to what is listed in RFCs 2369 and 2929.
841 Also, for "Auto-Submitted", treat anything other than "no" as
842 non-personal, in accordance with RFC 3834. (Previously it treated
843 anything starting "auto-" as non-personal.)
845 TF/01 The control=submission/name=... option had a problem with syntax
846 errors if the name included a slash character. The /name= option
847 now slurps the rest of the string, so it can include any characters
848 but it must come last in the list of options (after /sender_retain
851 PH/05 Some modifications to the interface to the fake nameserver for the new
859 TK/01 Added the "success_on_redirect" address verification option. See
860 NewStuff for rationale and an example.
862 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
864 PH/02 Patch to exigrep to allow it to work on syslog lines.
866 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
867 fread() to skip over the body file's header line, because in Cygwin the
868 header line is locked and is inaccessible.
870 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
871 co-exist for some time) to make it clear that it is the Exim ID that is
872 referenced, not the Message-ID: header line.
874 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
875 string_format() function, because snprintf() does not exist on all
878 PH/06 The use of forbid_filter_existstest now also locks out the use of the
879 ${stat: expansion item.
881 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
882 protocol synchronization error", to keep the pedants happy.
884 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
885 well as for IRIX systems, when gcc is being used. See the host.c source
888 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
890 PH/10 Named domain lists were not working if used in a queue_smtp_domains
893 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
894 transport and to the smtp transport in LMTP mode.
896 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
898 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
899 run a filter in a subprocess. This could lead to confusion in subsequent
900 lookups in the parent process. There should also be a search_tidyup() at
901 the end of the subprocess.
903 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
904 only if the host matched helo_try_verify_hosts, which caused the
905 verification to occur when the EHLO/HELO command was issued. The ACL just
906 tested the remembered result. Now, if a previous verification attempt has
907 not happened, "verify = helo" does it there and then.
909 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
911 TK/03 Fix log output including CR from clamd.
913 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
914 exist provoked a memory error which could cause a segfault.
916 PH/15 Installed PCRE 6.2
918 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
920 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
921 of the problem. Specifically, suggested +O2 rather than +O1 for the
924 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
926 PH/20 If a delivery was routed to a non-standard port by means of an SRV
927 record, the port was not correctly logged when the outgoing_port log
928 selector was set (it logged the transort's default port).
930 PH/21 Added support for host-specific ports to manualroute, queryprogram,
931 fallback_hosts, and "hosts" in the smtp transport.
933 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
934 host errors such as "Connection refused".
936 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
937 authentication with radiusclient 0.4.9:
939 - Error returned from rc_read_config was caught wrongly
940 - Username/password not passed on to radius server due to wrong length.
942 The presumption is that some radiusclient API changes for 4.51/PH/17
943 were not taken care of correctly. The code is still untested by me (my
944 Linux distribution still has 0.3.2 of radiusclient), but it was
945 contributed by a Radius user.
947 PH/24 When doing a callout, the value of $domain wasn't set correctly when
948 expanding the "port" option of the smtp transport.
950 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
951 while reading a MIME header. Thanks to Tom Hughes for a patch.
953 PH/24 Include config.h inside local_scan.h so that configuration settings are
956 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
957 that in an ACL for RCPT (for example), you can examine exactly what was
960 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
961 commands, but it was not correctly comparing the address with the actual
962 client host address. Thus, it would show the EHLO address in Received:
963 header lines when this was not necessary.
965 PH/27 Added the % operator to ${eval:}.
967 PH/28 Exim tries to create and chdir to its spool directory when it starts;
968 it should be ignoring failures (because with -C, for example, it has lost
969 privilege). It wasn't ignoring creation failures other than "already
972 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
975 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
976 because an input line was too long, either on its own, or by virtue of
977 too many continuations, the temporary file was not being removed, and the
978 return code was incorrect.
980 PH/31 Missing "BOOL" in function definition in filtertest.c.
982 PH/32 Applied Sieve patches from the maintainer.
984 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
986 PH/33 Added "verify = not_blind".
988 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
989 Local/Makefile (with some defaults set). These are used in built scripts
990 such as exicyclog, but they have never been used in the exim_install
991 script (though there are many overriding facilities there). I have
992 arranged that the exim_install script now takes note of these two
995 PH/35 Installed configuration files for Dragonfly.
997 PH/36 When a locally submitted message by a trusted user did not contain a
998 From: header, and the sender address was obtained from -f or from an SMTP
999 MAIL command, and the trusted user did not use -F to supply a sender
1000 name, $originator_name was incorrectly used when constructing a From:
1001 header. Furthermore, $originator_name was used for submission mode
1002 messages from external hosts without From: headers in a similar way,
1003 which is clearly wrong.
1005 PH/37 Added control=suppress_local_fixups.
1007 PH/38 When log_selector = +received_sender was set, and the addition of the
1008 sender made the log line's construction buffer exactly full, or one byte
1009 less than full, an overflow happened when the terminating "\n" was
1012 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1013 when the result of a list match is failure because a DNS lookup failed.
1015 PH/40 RM_COMMAND is now used in the building process.
1017 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1018 the "build-* directories that it finds.
1020 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1021 address in a domain literal was a prefix of an interface address.
1023 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1024 when verifying a sender address, unless rewrite_headers is false.
1026 PH/44 Wrote a long comment about why errors_to addresses are verified as
1027 recipients, not senders.
1029 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1030 the ratelimit ACL was added.
1032 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1034 PH/46 Added extra information about PostgreSQL errors to the error string.
1036 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1037 avoiding the need to install special zones in a real server. This is
1038 backwards compatible; if it can't find the fake resolver, it drops back.
1039 Thus, both old and new test suites can be run.
1041 TF/02 Added util/ratelimit.pl
1043 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1044 clock is set back in time.
1046 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1047 Candler <B.Candler@pobox.com>.
1049 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1050 OK for addresses that are the result of redirections.
1052 PH/48 A number of further additions for the benefit of the new test suite,
1053 including a fake gethostbyname() that interfaces to the fake DNS resolver
1056 TF/06 The fix for widen_domains has also been applied to qualify_single and
1057 search_parents which are the other dnslookup options that can cause
1060 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1063 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1065 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1066 service" if it encountered an MX record with an empty target hostname.
1067 The message is now "an MX or SRV record indicated no SMTP service".
1069 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1070 if the DNS of the sending site is misconfigured. This is quite a
1071 common situation. This change restores the behaviour of treating a
1072 helo verification defer as a failure.
1074 PH/51 If self=fail was set on a router, the bounce message did not include the
1075 actual error message.
1081 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1083 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1084 command itself ended in error, the underlying message about the transport
1085 filter timeout was being overwritten with the pipe command error. Now the
1086 underlying error message should be appended to the second error message.
1088 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1090 PH/02 Reduce the amount of output that "make" produces by default. Full output
1091 can still be requested.
1093 PH/03 The warning log line about a condition test deferring for a "warn" verb
1094 was being output only once per connection, rather than after each
1095 occurrence (because it was using the same function as for successful
1096 "warn" verbs). This seems wrong, so I have changed it.
1098 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1099 it should not have, which might have caused a crash in the right
1100 circumstances, but probably never did.
1102 PH/04 Installed a modified version of Tony Finch's patch to make submission
1103 mode fix the return path as well as the Sender: header line, and to
1104 add a /name= option so that you can make the user's friendly name appear
1107 TF/03 Added the control = fakedefer ACL modifier.
1109 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1110 Mark Lowes for thorough testing.
1112 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1114 TK/03 Merged latest SRS patch from Miles Wilton.
1116 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1117 with the definition in sysexits.h (which is #included earlier).
1118 Fortunately, Exim does not actually use EX_OK. The code used to try to
1119 preserve the sysexits.h value, by assumimg that macro definitions were
1120 scanned for macro replacements. I have been disabused of this notion,
1121 so now the code just undefines EX_OK before #including unistd.h.
1123 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1124 in the smtp transport. When a block could not be written in a single
1125 write() function, the timeout was being re-applied to each part-write.
1126 This seems wrong - if the receiver was accepting one byte at a time it
1127 would take for ever. The timeout is now adjusted when this happens. It
1128 doesn't have to be particularly precise.
1130 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1131 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1133 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1134 without a domain if the check to <postmaster@domain> fails.
1136 SC/01 Eximstats: added -xls and the ability to specify output files
1137 (patch written by Frank Heydlauf).
1139 SC/02 Eximstats: use FileHandles for outputing results.
1141 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1143 SC/04 Eximstats: fixed display of large numbers with -nvr option
1145 SC/05 Eximstats: fixed merging of reports with empty tables.
1147 SC/06 Eximstats: added the -include_original_destination flag
1149 SC/07 Eximstats: removed tabs and trailing whitespace.
1151 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1153 TK/06 MBOX spool code: Add real "From " MBOX separator line
1154 so the .eml file is really in mbox format (even though
1155 most programs do not really care). Patch from Alex Miller.
1157 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1158 The latter is generated from $received_to and is only set if the
1159 message has one envelope recipient. SA can use these headers,
1160 obviously out-of-the-box. Patch from Alex Miller.
1162 PH/08 The ${def test on a variable was returning false if the variable's
1163 value was "0", contrary to what the specification has always said!
1164 The result should be true unless the variable is empty.
1166 PH/09 The syntax error of a character other than { following "${if
1167 def:variable_name" (after optional whitespace) was not being diagnosed.
1168 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1169 accidental colon was present, for example, could give incorrect results.
1171 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1172 result is used (not including appendfile, where other changes are about
1175 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1176 This involved changing a lot of size variables from int to off_t. It
1177 should work with maildirs and everything.
1179 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1180 spamd dying while we are connected to it.
1182 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1185 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1186 function for BDB 4.3.
1188 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1190 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1191 the sources for PCRE have been re-arranged and more files are now
1194 PH/15 The code I had for printing potentially long long variables in PH/11
1195 above was not the best (it lost precision). The length of off_t variables
1196 is now inspected at build time, and an appropriate printing format (%ld
1197 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1198 to be "long long int" or "long int". This is needed for the internal
1199 formatting function string_vformat().
1201 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1202 the configuration file to be ":syslog", then the script "guesses" where
1203 the logs files are, rather than using the compiled in default. In our
1204 case the guess is not the same as the compiled default, so the script
1205 suddenly stopped working when I started to use syslog. The patch checks
1206 to see if log_file_path is "". If so, it attempts to read it from exim
1207 with no configuration file to get the compiled in version, before it
1208 falls back to the previous guessing code."
1210 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1211 implementing BATV in an Exim configuration. See NewStuff for the gory
1214 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1215 Makefile that are specific to HP-UX.
1217 PH/18 If the "use_postmaster" option was set for a recipient callout together
1218 with the "random" option, the postmaster address was used as the MAIL
1219 FROM address for the random test, but not for the subsequent recipient
1220 test. It is now used for both.
1222 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1223 patch removes a few documentation additions to RFC 3028, because the
1224 latest draft now contains them. It adds the new en;ascii-case comparator
1225 and a new error check for 8bit text in MIME parts. Comparator and
1226 require names are now matched exactly. I enabled the subaddress
1227 extension, but it is not well tested yet (read: it works for me)."
1229 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1230 rework some of the code of TK/09 above to avoid the hardwired use of
1231 "%lld" and "long long". Replaced the call to snprintf() with a call to
1234 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1235 records point to non-existent hosts", "retry timeout exceeded", and
1236 "retry time not reached for any host after a long failure period".
1238 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1239 experimental DomainKeys support:
1241 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1242 (2) On an error such as an illegally used "control", the wrong name for
1243 the control was given.
1245 These problems did NOT occur unless DomainKeys support was compiled.
1247 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1249 PH/24 Added ${if match_ip condition.
1251 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1252 databases so that it will be absolutely obvious if a crash occurs in the
1253 DB library. This is a regular occurrence (often caused by mis-matched
1256 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1257 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1258 picked up on a user's system that detects such things. There doesn't seem
1259 to be a gcc warning option for this - only an attribute that has to be
1260 put on the function's prototype. It seems that in Fedora Core 4 they have
1261 set this on a number of new functions. No doubt there will be more in due
1264 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1265 specify a transport. However, if an address that was verified by such a
1266 router was the subject of a callout, Exim crashed because it tried to
1267 read the rcpt_include_affixes from the non-existent transport. Now it
1268 just assumes that the setting of that option is false. This bug was
1269 introduced by 4.51/PH/31.
1271 PH/28 Changed -d+all to exclude +memory, because that information is very
1272 rarely of interest, but it makes the output a lot bigger. People tend to
1273 do -d+all out of habit.
1275 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1276 code in os-type was giving problems when libc.so lives in lib64, like on
1279 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1280 aren't the modern standard, and it seems that some systems' include files
1281 don't always have them. Exim was already checking for some of the newer
1282 ones like T_AAAA, and defining it itself. I've added checks for all the
1283 record types that Exim uses.
1285 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1286 not automatically generating a new one, as it is supposed to. This
1287 prevented TLS from working. If the file did exist, but contained invalid
1288 data, a new version was generated, as expected. It was only the case of a
1289 non-existent file that was broken.
1291 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1292 with a change in libdomainkeys > 0.64.
1294 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1295 from DNS. If the selector record carries the flag, it now has
1296 precedence over the domain-wide flag.
1298 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1300 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1301 the use of a port name that isn't defined in /etc/services) occurred, the
1302 message was deferred as in a normal delivery, and thus remained on the
1303 spool, instead of being failed because of the mua_wrapper setting. This
1304 is now fixed, and I tidied up some of the mua_wrapper messages at the
1307 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1308 the messages in a hash of arrays rather than using individual hashes.
1309 This is a bit cleaner and results in dramatic memory savings, albeit
1310 at a slight CPU cost.
1312 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1313 as requested by Marc Sherman.
1315 SC/10 Eximstats: added histograms for user specified patterns as requested
1318 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1320 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1321 fopen() in the content-scanning modules that did not already have it.
1327 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1328 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1330 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1332 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1334 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1335 to test Sieve filters that use "vacation".
1337 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1338 that changes the way the GnuTLS parameters are stored in the cache file.
1339 The new format can be generated externally. For backward compatibility,
1340 if the data in the cache doesn't make sense, Exim assumes it has read an
1341 old-format file, and it generates new data and writes a new file. This
1342 means that you can't go back to an older release without removing the
1345 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1346 work if there are any delivery delays because "one_time" forces the
1347 parent to be marked "delivered", so its unseen clone is never tried
1348 again. For this reason, Exim now forbids the simultaneous setting of
1351 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1352 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1353 entry if you want to know the details. The fix, however, neglected to
1354 consider the case when local delivery batching is involved. The test for
1355 "previously delivered" was not happening when checking to see if an
1356 address could be batched with a previous (undelivered) one; under
1357 certain circumstances this could lead to multiple deliveries to the same
1360 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1361 in its include files, and this causes problems building Exim.
1363 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1364 header_syntax) but Exim was just ignoring anything given after a slash.
1365 In particular, this caused confusion with an attempt to use "verify =
1366 reverse_host_lookup/defer_ok". An error is now given when options are
1367 supplied for verify items that do not have them. (Maybe reverse_host_
1368 lookup should have a defer_ok option, but that's a different point.)
1370 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1371 defined by RFC 821) to 2048, because there were problems with some AUTH
1372 commands, and RFC 1869 says the size should be increased for extended
1373 SMTP commands that take arguments.
1375 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1378 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1379 "unknown" error; now it says that the functionality isn't in the binary.
1381 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1382 an address' error message when a string expansion fails (syntax or
1383 whatever). Otherwise the password may appear in the log. Following change
1384 PH/42 below, there is no longer a chance of it appearing in a bounce
1387 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1389 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1390 of its addresses was ever tried. (Bugzilla bug #2.)
1392 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1393 the result incorrectly in the debug output. (It correctly added a newline
1394 to what was transported.)
1396 TF/01 Added $received_time.
1398 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1399 commented out examples of how to interface to a virus scanner and to
1400 SpamAssassin. Also added commented examples of av_scanner and
1401 spamd_address settings.
1403 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1404 and controls are allowed in which ACLs. There were a couple of minor
1405 errors. Some of the entries in the conditions table (which is a table of
1406 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1407 negation of where the condition IS allowed.
1409 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1411 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1412 header file does not have a version number, so I've had to invent a new
1413 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1414 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1415 radiusclient), but it was contributed by a Radius user.
1417 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1418 files or directories, for OpenSSL.
1420 PH/19 When an Exim process that is running as root has to create an Exim log
1421 file, it does so in a subprocess that runs as exim:exim so as to get the
1422 ownership right at creation (otherwise, other Exim processes might see
1423 the file with the wrong ownership). There was no test for failure of this
1424 fork() call, which would lead to the process getting stuck as it waited
1425 for a non-existent subprocess. Forks do occasionally fail when resources
1426 run out. I reviewed all the other calls to fork(); they all seem to check
1429 PH/20 When checking for unexpected SMTP input at connect time (before writing
1430 the banner), Exim was not dealing correctly with a non-positive return
1431 from the read() function. If the client had disconnected by this time,
1432 the result was a log entry for a synchronization error with an empty
1433 string after "input=" when read() returned zero. If read() returned -1
1434 (an event I could not check), uninitialized data bytes were printed.
1435 There were reports of junk text (parts of files, etc) appearing after
1438 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1440 PH/22 Added support for macro redefinition, and (re)definition in between
1441 driver and ACL definitions.
1443 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1444 forgetting to use the resulting value; it was using the unexpanded value.
1446 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1447 hadn't been configured. The fix is from Juergen Kreileder, who
1448 understands it better than I do:
1450 "Here's what I see happening with three configured cyrus_sasl
1451 authenticators configured (plain, login, cram-md5):
1453 On startup auth_cyrus_sasl_init() gets called for each of these.
1454 This means three calls to sasl_listmech() without a specified mech_list.
1455 => SASL tests which mechs of all available mechs actually work
1456 => three warnings about OTP not working
1457 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1459 With the patch, sasl_listmech() also gets called three times. But now
1460 SASL's mech_list option is set to the server_mech specified in the the
1461 authenticator. Or in other words, the answer from sasl_listmech()
1462 gets limited to just the mech you're testing for (which is different
1464 => the return list contains just 'plain' or 'login', 'cram-md5' or
1465 nothing depending on the value of ob->server_mech.
1467 I've just tested the patch: Authentication still works fine,
1468 unavailable mechs specified in the exim configuration are still
1469 caught, and the auth.log warnings about OTP are gone."
1471 PH/25 When debugging is enabled, the contents of the command line are added
1472 to the debugging output, even when log_selector=+arguments is not
1475 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1476 answer is "GNU", and only if the return is "GNU/something" is the answer
1479 PH/27 $acl_verify_message is now set immediately after the failure of a
1480 verification in an ACL, and so is available in subsequent modifiers. In
1481 particular, the message can be preserved by coding like this:
1483 warn !verify = sender
1484 set acl_m0 = $acl_verify_message
1486 Previously, $acl_verify_message was set only while expanding "message"
1487 and "log_message" when a very denied access.
1489 PH/28 Modified OS/os.c-Linux with
1491 -#ifndef OS_LOAD_AVERAGE
1492 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1494 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1495 nomenclature these days.)
1497 PH/29 Installed patch from the Sieve maintainer that adds the options
1498 sieve_useraddress and sieve_subaddress to the redirect router.
1500 PH/30 In these circumstances:
1501 . Two addresses routed to the same list of hosts;
1502 . First host does not offer TLS;
1503 . First host accepts first address;
1504 . First host gives temporary error to second address;
1505 . Second host offers TLS and a TLS session is established;
1506 . Second host accepts second address.
1507 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1508 and peerdn, if requested) that were in fact used only for the second
1511 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1512 attention to any local part prefix or suffix that was matched by the
1513 router that accepted the address. It now behaves in the same way as it
1514 does for delivery: the affixes are removed from the local part unless
1515 rcpt_include_affixes is set on the transport.
1517 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1518 timeout during the DATA phase of an incoming message.
1520 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1521 applied a patch sanctioned by the Sieve maintainer.
1523 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1524 the uid or gid is negative. A case of a negative gid caused this to be
1525 noticed. The fix allows for either to be negative.
1527 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1528 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1531 PH/36 The change PH/12 above was broken. Fixed it.
1533 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1534 the grounds that routing the same address twice would always produce the
1535 same answer. This might have been true once, but it is certainly no
1536 longer true now. Routing a child address may depend on the previous
1537 routing that produced that child. Some complicated redirection strategies
1538 went wrong when messages had multiple recipients, and made Exim's
1539 behaviour dependent on the order in which the addresses were given.
1541 I have moved the duplicate checking until after the routing is complete.
1542 Exim scans the addresses that are assigned to local and remote
1543 transports, and removes any duplicates. This means that more work will be
1544 done, as duplicates will always all be routed, but duplicates are
1545 presumably rare, so I don't expect this is of any significance.
1547 For deliveries to pipes, files, and autoreplies, the duplicate checking
1548 still happens during the routing process, since they are not going to be
1551 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1552 It corrects a timeout issue with spamd. This is Ian's comment: "The
1553 background is that sometimes spamd either never reads data from a
1554 connection it has accepted, or it never writes response data. The exiscan
1555 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1556 blindly assumes that writes won't block so it may never time out."
1558 PH/39 Allow G after quota size as well as K and M.
1560 PH/40 The value set for $authenticated_id in an authenticator may not contain
1561 binary zeroes or newlines because the value is written to log lines and
1562 to spool files. There was no check on this. Now the value is run through
1563 the string_printing() function so that such characters are converted to
1564 printable escape sequences.
1566 PH/41 $message_linecount is a new variable that contains the total number of
1567 lines in the message. Compare $body_linecount, which is the count for the
1570 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1571 bounce and delay warning messages, except in certain special cases, which
1574 (a) An SMTP error message from a remote host;
1575 (b) A message specified in a :fail: redirection;
1576 (c) A message specified in a "fail" command in a system filter;
1577 (d) A message specified in a FAIL return from the queryprogram router;
1578 (e) A message specified by the cannot_route_message router option.
1580 In these cases only, Exim does include the error details in bounce and
1581 warning messages. There are also a few cases where bland messages such
1582 as "unrouteable address" or "local delivery error" are given.
1584 PH/43 $value is now also set for the "else" part of a ${run expansion.
1586 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1587 being worked on, but at least Exim now implements the latest version to
1590 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1591 process to complete was treated as a delivery failure, a timeout while
1592 writing the message to the pipe was logged, but erroneously treated as a
1593 successful delivery. Such timeouts include transport filter timeouts. For
1594 consistency with the overall process timeout, these timeouts are now
1595 treated as errors, giving rise to delivery failures by default. However,
1596 there is now a new Boolean option for the pipe transport called
1597 timeout_defer, which, if set TRUE, converts the failures into defers for
1598 both kinds of timeout. A transport filter timeout is now identified in
1601 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1602 systems where "make" and "gmake" are different, calling "gmake" at top
1603 level broke things. I've arranged for the value of $(MAKE) to be passed
1604 from the Makefile to this script so that it can call the same version of
1608 A note about Exim versions 4.44 and 4.50
1609 ----------------------------------------
1611 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1612 changes of various kinds. As a consequence, a big documentation update was
1613 needed. This delayed the release for rather longer than seemed good, especially
1614 in the light of a couple of (minor) security issues. Therefore, the changes
1615 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1616 release. So 4.44 and 4.50 are in effect two different branches that both start
1619 I have left the 4.50 change log unchanged; it contains all the changes since
1620 4.43. The change log for 4.44 is below; many of its items are identical to
1621 those for 4.50. This seems to be the most sensible way to preserve the
1622 historical information.
1628 1. Minor wording change to the doc/README.SIEVE file.
1630 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1631 computation of the current number of files was incorrect.
1633 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1634 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1637 4. Give more explanation in the error message when the command for a transport
1638 filter fails to execute.
1640 5. There are several places where Exim runs a non-Exim command in a
1641 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1642 was being done only for the command run by the queryprogram router. It is
1643 now done for all such subprocesses. The other cases are: ${run, transport
1644 filters, and the commands run by the lmtp and pipe transports.
1646 6. Added CONFIGURE_GROUP build-time option.
1648 7. Some older OS have a limit of 256 on the maximum number of file
1649 descriptors. Exim was using setrlimit() to set 1000 as a large value
1650 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1651 systems. I've change it so that if it can't get 1000, it tries for 256.
1653 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1654 was an oversight, and furthermore, ever since the addition of extra
1655 controls (e.g. 4.43/32), the checks on when to allow different forms of
1656 "control" were broken. There should now be diagnostics for all cases when a
1657 control that does not make sense is encountered.
1659 9. Added the /retain_sender option to "control=submission".
1661 10. $recipients is now available in the predata ACL (oversight).
1663 11. Tidy the search cache before the fork to do a delivery from a message
1664 received from the command line. Otherwise the child will trigger a lookup
1665 failure and thereby defer the delivery if it tries to use (for example) a
1666 cached ldap connection that the parent has called unbind on.
1668 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1669 of $address_data from the recipient verification was clobbered by the
1670 sender verification.
1672 13. The value of address_data from a sender verification is now available in
1673 $sender_address_data in subsequent conditions in the ACL statement.
1675 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1677 15. Added a new option "connect=<time>" to callout options, to set a different
1680 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1681 was its contents. (It was OK if the option was not defined at all.)
1683 17. A "Completed" log line is now written for messages that are removed from
1684 the spool by the -Mrm option.
1686 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1687 information about exactly what failed.
1689 19. Added -dd to debug only the daemon process.
1691 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1692 handles timeouts, both on the server side and network timeouts. Renamed the
1693 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1695 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1696 It is now set to "smtps".
1698 22. $host_address is now set to the target address during the checking of
1699 ignore_target_hosts.
1701 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1702 being passed; this would have caused $sender_host_name to have been used if
1703 matching the list had actually called for a host name (not very likely,
1704 since this list is usually IP addresses). A host name is now passed as
1707 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1708 code that specifies a non-restarting handler (typically sigaction() in
1709 modern systems) in an attempt to fix a rare and obscure crash bug.
1711 25. Narrowed the window for a race in the daemon that could cause it to ignore
1712 SIGCHLD signals. This is not a major problem, because they are used only to
1713 wake it up if nothing else does.
1715 26. A malformed maildirsize file could cause Exim to calculate negative values
1716 for the mailbox size or file count. Odd effects could occur as a result.
1717 The maildirsize information is now recalculated if the size or filecount
1720 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1721 support for a long time. Removed HAVE_SYS_VFS_H.
1723 28. Installed the latest version of exipick from John Jetmore.
1725 29. In an address list, if the pattern was not a regular expression, an empty
1726 subject address (from a bounce message) matched only if the pattern was an
1727 empty string. Non-empty patterns were not even tested. This was the wrong
1728 because it is perfectly reasonable to use an empty address as part of a
1729 database query. An empty address is now tested by patterns that are
1730 lookups. However, all the other forms of pattern expect the subject to
1731 contain a local part and a domain, and therefore, for them, an empty
1732 address still always fails if the pattern is not itself empty.
1734 30. Exim went into a mad DNS loop when attempting to do a callout where the
1735 host was specified on an smtp transport, and looking it up yielded more
1736 than one IP address.
1738 31. Re-factored the code for checking spool and log partition space into a
1739 function that finds that data and another that does the check. The former
1740 is then used to implement four new variables: $spool_space, $log_space,
1741 $spool_inodes, and $log_inodes.
1743 32. The RFC2047 encoding function was originally intended for short strings
1744 such as real names; it was not keeping to the 75-character limit for
1745 encoded words that the RFC imposes. It now respects the limit, and
1746 generates multiple encoded words if necessary. To be on the safe side, I
1747 have increased the buffer size for the ${rfc2047: expansion operator from
1750 33. It is now permitted to omit both strings after an "if" condition; if the
1751 condition is true, the result is "true". As before, when the second string
1752 is omitted, a false condition yields an empty string. This makes it less
1753 cumbersome to write custom ACL and router conditions.
1755 34. Failure to deliver a bounce message always caused it to be frozen, even if
1756 there was an errors_to setting on the router. The errors_to setting is now
1759 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1760 canonical form (fully expanded) before being placed in
1761 $sender_host_address.
1763 36. The table in the code that translates DNS record types into text (T_A to
1764 "A" for instance) was missing entries for NS and CNAME. It is just possible
1765 that this could have caused confusion if both these types were looked up
1766 for the same domain, because the text type is used as part of Exim's
1767 per-process caching. But the chance of anyone hitting this buglet seems
1770 37. The dnsdb lookup has been extended in a number of ways.
1772 (1) There is a new type, "zns", which walks up the domain tree until it
1773 finds some nameserver records. It should be used with care.
1775 (2) There is a new type, "mxh", which is like "mx" except that it returns
1776 just the host names, not the priorities.
1778 (3) It is now possible to give a list of domains (or IP addresses) to be
1779 looked up. The behaviour when one of the lookups defers can be
1780 controlled by a keyword.
1782 (4) It is now possible to specify the separator character for use when
1783 multiple records are returned.
1785 38. The dnslists ACL condition has been extended: it is now possible to supply
1786 a list of IP addresses and/or domains to be looked up in a particular DNS
1789 39. Added log_selector=+queue_time_overall.
1791 40. When running the queue in the test harness, wait just a tad after forking a
1792 delivery process, to get repeatability of debugging output.
1794 41. Include certificate and key file names in error message when GnuTLS fails
1795 to set them up, because the GnuTLS error message doesn't include the name
1796 of the failing file when there is a problem reading it.
1798 42. Allow both -bf and -bF in the same test run.
1800 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1802 44. The "Exiscan patch" is now merged into the mainline Exim source.
1804 45. Sometimes the final signoff response after QUIT could fail to get
1805 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1806 < 0 before doing a fflush(). This bug looks as though it goes back to the
1807 introduction of TLS in release 3.20, but "sometimes" must have been rare
1808 because the tests only now provoked it.
1810 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1811 (this can affect the format of dates).
1813 47. exim_tidydb, when checking for the continued existence of a message for
1814 which it has found a message-specific retry record, was not finding
1815 messages that were in split spool directories. Consequently, it was
1816 deleting retry records that should have stayed in existence.
1818 48. Steve fixed some bugs in eximstats.
1820 49. The SPA authentication driver was not abandoning authentication and moving
1821 on to the next authenticator when an expansion was forced to fail,
1822 contradicting the general specification for all authenticators. Instead it
1823 was generating a temporary error. It now behaves as specified.
1825 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1826 (the order specifies the preference for clients). The order is now AES256,
1827 AES128, 3DES, ARCFOUR128.
1829 51. Small patch to Sieve code - explicitly set From: when generating an
1832 52. Exim crashed if a remote delivery caused a very long error message to be
1833 recorded - for instance if somebody sent an entire SpamAssassin report back
1834 as a large number of 550 error lines. This bug was coincidentally fixed by
1835 increasing the size of one of Exim's internal buffers (big_buffer) that
1836 happened as part of the Exiscan merge. However, to be on the safe side, I
1837 have made the code more robust (and fixed the comments that describe what
1840 53. Now that there can be additional text after "Completed" in log lines (if
1841 the queue_time_overall log selector is set), a one-byte patch to exigrep
1842 was needed to allow it to recognize "Completed" as not the last thing in
1845 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1846 patch that reportedly fixes this has been added. I am not expert enough to
1847 create a test for it. This is what the patch creator wrote:
1849 "I found a little strange behaviour of ldap code when working with
1850 Windows 2003 AD Domain, where users was placed in more than one
1851 Organization Units. When I tried to give exim partial DN, the exit code
1852 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1853 But simultaneously result of request was absolutely normal ldap result,
1854 so I produce this patch..."
1856 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1857 REFERENCE, so I have modified the code to exclude the patch when that macro
1860 55. Some experimental protocols are using DNS PTR records for new purposes. The
1861 keys for these records are domain names, not reversed IP addresses. The
1862 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1863 leaves it alone. Component reversal etc. now happens only for IP addresses.
1866 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1868 57. Double the size of the debug message buffer (to 2048) so that more of very
1869 long debug lines gets shown.
1871 58. The exicyclog utility now does better if the number of log files to keep
1872 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1874 59. Two changes related to the smtp_active_hostname option:
1876 (1) $smtp_active_hostname is now available as a variable.
1877 (2) The default for smtp_banner uses $smtp_active_hostname instead
1878 of $primary_hostname.
1880 60. The host_aton() function is supposed to be passed a string that is known
1881 to be a valid IP address. However, in the case of IPv6 addresses, it was
1882 not checking this. This is a hostage to fortune. Exim now panics and dies
1883 if the condition is not met. A case was found where this could be provoked
1884 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1885 components; fortuitously, this particular loophole had already been fixed
1886 by change 4.50/55 above.
1888 If there are any other similar loopholes, the new check in host_aton()
1889 itself should stop them being exploited. The report I received stated that
1890 data on the command line could provoke the exploit when Exim was running as
1891 exim, but did not say which command line option was involved. All I could
1892 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1893 running as the user.
1896 61. There was a buffer overflow vulnerability in the SPA authentication code
1897 (which came originally from the Samba project). I have added a test to the
1898 spa_base64_to_bits() function which I hope fixes it.
1901 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1902 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1904 63. The daemon start-up calls getloadavg() while still root for those OS that
1905 need the first call to be done as root, but it missed one case: when
1906 deliver_queue_load_max is set with deliver_drop_privilege. This is
1907 necessary for the benefit of the queue runner, because there is no re-exec
1908 when deliver_drop_privilege is set.
1910 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1911 This has been fixed.
1913 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1914 was in use, was not putting the data itself into the right store pool;
1915 consequently, it could be overwritten for a subsequent message in the same
1916 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1919 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1921 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1922 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1923 the function were treating the return as a boolean value, which happened to
1924 work because 0=false and not-0=true, but is not correct code.
1926 68. The host_aton() function was not handling scoped IPv6 addresses (those
1927 with, for example, "%eth0" on the end) correctly.
1929 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1930 negated items (that is, ~something) in unsigned ints. Some compilers
1931 apparently mutter when there is no cast.
1933 70. If an address verification called from an ACL failed, and did not produce a
1934 user-specific message (i.e. there was only a "system" message), nothing was
1935 put in $acl_verify_message. In this situation, it now puts the system
1938 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1939 session; change 4.31/43 added the unwanted input to the log line - except
1940 that it did not do this in the start of session case. It now does.
1942 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1943 This is wrong and can cause the other end to generate a synchronization
1944 error if it is another Exim or anything else that does the synchronization
1945 check. A QUIT command is no longer sent after a timeout.
1947 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1948 during host lookups.
1950 74. The defer_ok option of callout verification was not working if it was used
1951 when verifying addresses in header lines, that is, for this case:
1953 verify = header_sender/callout=defer_ok
1955 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1956 those file descriptors could be used for SMTP connections. If anything
1957 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1958 could be sent to the SMTP client, causing chaos. The daemon now opens
1959 stdin, stdout, and stderr to /dev/null when it puts itself into the
1962 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1963 log by default. The user can override this with suitable Perl magic.
1965 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1966 the log message when discard triggers, was not working for the DATA ACL or
1967 for the non-SMTP ACL.
1969 78. Error message wording change in sieve.c.
1971 79. If smtp_accept_max_per_host was set, the number of connections could be
1972 restricted to fewer than expected, because the daemon was trying to set up
1973 a new connection before checking whether the processes handling previous
1974 connections had finished. The check for completed processes is now done
1975 earlier. On busy systems, this bug wouldn't be noticed because something
1976 else would have woken the daemon, and it would have reaped the completed
1979 80. If a message was submitted locally by a user whose login name contained one
1980 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1981 It caused a spool format error. I have fixed the spool reading code. A
1982 related problem was that the "from" clause in the Received: line became
1983 illegal because of the space(s). It is now covered by ${quote_local_part.
1985 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1988 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1990 83. Patch from Sieve maintainer to fix unterminated string problem in
1991 "vacation" handling.
1993 84. Some minor changes to the Linux configuration files to help with other
1994 OS variants using glibc.
1996 85. One more patch for Sieve to update vacation handling to latest spec.
1999 ----------------------------------------------------
2000 See the note above about the 4.44 and 4.50 releases.
2001 ----------------------------------------------------
2007 1. Change 4.43/35 introduced a bug that caused file counts to be
2008 incorrectly computed when quota_filecount was set in an appendfile
2011 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2012 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2015 3. Give more explanation in the error message when the command for a transport
2016 filter fails to execute.
2018 4. There are several places where Exim runs a non-Exim command in a
2019 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2020 was being done only for the command run by the queryprogram router. It is
2021 now done for all such subprocesses. The other cases are: ${run, transport
2022 filters, and the commands run by the lmtp and pipe transports.
2024 5. Some older OS have a limit of 256 on the maximum number of file
2025 descriptors. Exim was using setrlimit() to set 1000 as a large value
2026 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2027 systems. I've change it so that if it can't get 1000, it tries for 256.
2029 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2030 was an oversight, and furthermore, ever since the addition of extra
2031 controls (e.g. 4.43/32), the checks on when to allow different forms of
2032 "control" were broken. There should now be diagnostics for all cases when a
2033 control that does not make sense is encountered.
2035 7. $recipients is now available in the predata ACL (oversight).
2037 8. Tidy the search cache before the fork to do a delivery from a message
2038 received from the command line. Otherwise the child will trigger a lookup
2039 failure and thereby defer the delivery if it tries to use (for example) a
2040 cached ldap connection that the parent has called unbind on.
2042 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2043 of $address_data from the recipient verification was clobbered by the
2044 sender verification.
2046 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2047 was its contents. (It was OK if the option was not defined at all.)
2049 11. A "Completed" log line is now written for messages that are removed from
2050 the spool by the -Mrm option.
2052 12. $host_address is now set to the target address during the checking of
2053 ignore_target_hosts.
2055 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2056 being passed; this would have caused $sender_host_name to have been used if
2057 matching the list had actually called for a host name (not very likely,
2058 since this list is usually IP addresses). A host name is now passed as
2061 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2062 code that specifies a non-restarting handler (typically sigaction() in
2063 modern systems) in an attempt to fix a rare and obscure crash bug.
2065 15. Narrowed the window for a race in the daemon that could cause it to ignore
2066 SIGCHLD signals. This is not a major problem, because they are used only to
2067 wake it up if nothing else does.
2069 16. A malformed maildirsize file could cause Exim to calculate negative values
2070 for the mailbox size or file count. Odd effects could occur as a result.
2071 The maildirsize information is now recalculated if the size or filecount
2074 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2075 support for a long time. Removed HAVE_SYS_VFS_H.
2077 18. Updated exipick to current release from John Jetmore.
2079 19. Allow an empty sender to be matched against a lookup in an address list.
2080 Previously the only cases considered were a regular expression, or an
2083 20. Exim went into a mad DNS lookup loop when doing a callout where the
2084 host was specified on the transport, if the DNS lookup yielded more than
2087 21. The RFC2047 encoding function was originally intended for short strings
2088 such as real names; it was not keeping to the 75-character limit for
2089 encoded words that the RFC imposes. It now respects the limit, and
2090 generates multiple encoded words if necessary. To be on the safe side, I
2091 have increased the buffer size for the ${rfc2047: expansion operator from
2094 22. Failure to deliver a bounce message always caused it to be frozen, even if
2095 there was an errors_to setting on the router. The errors_to setting is now
2098 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2099 canonical form (fully expanded) before being placed in
2100 $sender_host_address.
2102 24. Updated eximstats to version 1.33
2104 25. Include certificate and key file names in error message when GnuTLS fails
2105 to set them up, because the GnuTLS error message doesn't include the name
2106 of the failing file when there is a problem reading it.
2108 26. Expand error message when OpenSSL has problems setting up cert/key files.
2111 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2112 (this can affect the format of dates).
2114 28. exim_tidydb, when checking for the continued existence of a message for
2115 which it has found a message-specific retry record, was not finding
2116 messages that were in split spool directories. Consequently, it was
2117 deleting retry records that should have stayed in existence.
2119 29. eximstats updated to version 1.35
2120 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2121 1.35 - bugfix such that pie charts by volume are generated correctly
2123 30. The SPA authentication driver was not abandoning authentication and moving
2124 on to the next authenticator when an expansion was forced to fail,
2125 contradicting the general specification for all authenticators. Instead it
2126 was generating a temporary error. It now behaves as specified.
2128 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2129 (the order specifies the preference for clients). The order is now AES256,
2130 AES128, 3DES, ARCFOUR128.
2132 31. Small patch to Sieve code - explicitly set From: when generating an
2135 32. Exim crashed if a remote delivery caused a very long error message to be
2136 recorded - for instance if somebody sent an entire SpamAssassin report back
2137 as a large number of 550 error lines. This bug was coincidentally fixed by
2138 increasing the size of one of Exim's internal buffers (big_buffer) that
2139 happened as part of the Exiscan merge. However, to be on the safe side, I
2140 have made the code more robust (and fixed the comments that describe what
2143 33. Some experimental protocols are using DNS PTR records for new purposes. The
2144 keys for these records are domain names, not reversed IP addresses. The
2145 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2146 leaves it alone. Component reversal etc. now happens only for IP addresses.
2149 34. The host_aton() function is supposed to be passed a string that is known
2150 to be a valid IP address. However, in the case of IPv6 addresses, it was
2151 not checking this. This is a hostage to fortune. Exim now panics and dies
2152 if the condition is not met. A case was found where this could be provoked
2153 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2154 components; fortuitously, this particular loophole had already been fixed
2155 by change 4.50/55 or 4.44/33 above.
2157 If there are any other similar loopholes, the new check in host_aton()
2158 itself should stop them being exploited. The report I received stated that
2159 data on the command line could provoke the exploit when Exim was running as
2160 exim, but did not say which command line option was involved. All I could
2161 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2162 running as the user.
2165 35. There was a buffer overflow vulnerability in the SPA authentication code
2166 (which came originally from the Samba project). I have added a test to the
2167 spa_base64_to_bits() function which I hope fixes it.
2170 36. The daemon start-up calls getloadavg() while still root for those OS that
2171 need the first call to be done as root, but it missed one case: when
2172 deliver_queue_load_max is set with deliver_drop_privilege. This is
2173 necessary for the benefit of the queue runner, because there is no re-exec
2174 when deliver_drop_privilege is set.
2176 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2177 was in use, was not putting the data itself into the right store pool;
2178 consequently, it could be overwritten for a subsequent message in the same
2179 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2182 38. Sometimes the final signoff response after QUIT could fail to get
2183 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2184 < 0 before doing a fflush(). This bug looks as though it goes back to the
2185 introduction of TLS in release 3.20, but "sometimes" must have been rare
2186 because the tests only now provoked it.
2192 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2193 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2194 Now it returns an integer. A number of calls were still expecting a T/F
2195 return. Fortuitously, in all cases, the tests worked in OK situations,
2196 which is the norm. However, things would have gone wrong on any write
2197 failures on the smtp file descriptor. This function is used when sending
2198 messages over SMTP and also when doing verify callouts.
2200 2. When Exim is called to do synchronous delivery of a locally submitted
2201 message (the -odf or -odi options), it no longer closes stderr before doing
2204 3. Implemented the mua_wrapper option.
2206 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2208 5. Implemented the functions header_remove(), header_testname(),
2209 header_add_at_position(), and receive_remove_recipient(), and exported them
2212 6. If an ACL "warn" statement specified the addition of headers, Exim already
2213 inserted X-ACL-Warn: at the start if there was no header name. However, it
2214 was not making this test for the second and subsequent header lines if
2215 there were newlines in the string. This meant that an invalid header could
2216 be inserted if Exim was badly configured.
2218 7. Allow an ACL "warn" statement to add header lines at the start or after all
2219 the Received: headers, as well as at the end.
2221 8. Added the rcpt_4xx retry error code.
2223 9. Added postmaster_mailfrom=xxx to callout verification option.
2225 10. Added mailfrom=xxxx to the callout verification option, for verify=
2228 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2229 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2231 12. Inserted some casts to stop certain compilers warning when using pointer
2232 differences as field lengths or precisions in printf-type calls (mostly
2233 affecting debugging statements).
2235 13. Added optional readline() support for -be (dynamically loaded).
2237 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2238 same clock tick as a message's arrival, so that its received time was the
2239 same as the "first fail" time on the retry record, and that message
2240 remained on the queue past the ultimate address timeout, every queue runner
2241 would try a delivery (because it was past the ultimate address timeout) but
2242 after another failure, the ultimate address timeout, which should have then
2243 bounced the address, did not kick in. This was a "< instead of <=" error;
2244 in most cases the first failure would have been in the next clock tick
2245 after the received time, and all would be well.
2247 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2248 being recognized when the domain list was tested by the match_domain
2249 condition in an expansion string.
2251 16. Added the ${str2b64: operator.
2253 17. Exim was always calling setrlimit() to set a large limit for the number of
2254 processes, without checking whether the existing limit was already
2255 adequate. (It did check for the limit on file descriptors.) Furthermore,
2256 errors from getrlimit() and setrlimit() were being ignored. Now they are
2257 logged to the main and panic logs, but Exim does carry on, to try to do its
2258 job under whatever limits there are.
2260 18. Imported PCRE 5.0.
2262 19. Trivial typo in log message " temporarily refused connection" (the leading
2265 20. If the log selector return_path_on_delivery was set and an address was
2266 redirected to /dev/null, the delivery process crashed because it assumed
2267 that a return path would always be set for a "successful" delivery. In this
2268 case, the whole delivery is bypassed as an optimization, and therefore no
2271 21. Internal re-arrangement: the function for sending a challenge and reading
2272 a response while authentication was assuming a zero-terminated challenge
2273 string. It's now changed to take a pointer and a length, to allow for
2274 binary data in such strings.
2276 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2278 23. Exim was not respecting finduser_retries when seeking the login of the
2279 uid under which it was called; it was always trying 10 times. (The default
2280 setting of finduser_retries is zero.) Also, it was sleeping after the final
2281 failure, which is pointless.
2283 24. Implemented tls_on_connect_ports.
2285 25. Implemented acl_smtp_predata.
2287 26. If the domain in control=submission is set empty, Exim assumes that the
2288 authenticated id is a complete email address when it generates From: or
2289 Sender: header lines.
2291 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2292 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2293 chown and chgrp in /bin and hostname in /usr/bin.
2295 28. Exim was keeping the "process log" file open after each use, just as it
2296 does for the main log. This opens the possibility of it remaining open for
2297 long periods when the USR1 signal hits a daemon. Occasional processlog
2298 errors were reported, that could have been caused by this. Anyway, it seems
2299 much more sensible not to leave this file open at all, so that is what now
2302 29. The long-running daemon process does not normally write to the log once it
2303 has entered its main loop, and it closes the log before doing so. This is
2304 so that log files can straightforwardly be renamed and moved. However,
2305 there are a couple of unusual error situations where the daemon does write
2306 log entries, and I had neglected to close the log afterwards.
2308 30. The text of an SMTP error response that was received during a remote
2309 delivery was being truncated at 512 bytes. This is too short for some of
2310 the long messages that one sometimes sees. I've increased the limit to
2313 31. It is now possible to make retry rules that apply only when a message has a
2314 specific sender, in particular, an empty sender.
2316 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2317 it possible to be selective about when SMTP synchronization is enforced.
2319 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2321 32. Implemented hosts_connection_nolog.
2323 33. Added an ACL for QUIT.
2325 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2328 35. Added mailbox_size and mailbox_filecount to appendfile.
2330 36. Added control = no_multiline_responses to ACLs.
2332 37. There was a bug in the logic of the code that waits for the clock to tick
2333 in the case where the clock went backwards by a substantial amount such
2334 that the microsecond fraction of "now" was more than the microsecond
2335 fraction of "then" (but the whole seconds number was less).
2337 38. Added support for the libradius Radius client library this is found on
2338 FreeBSD (previously only the radiusclient library was supported).
2344 1. When certain lookups returned multiple values in the form name=value, the
2345 quoting of the values was not always being done properly. Specifically:
2346 (a) If the value started with a double quote, but contained no whitespace,
2348 (b) If the value contained whitespace other than a space character (i.e.
2349 tabs or newlines or carriage returns) it was not quoted.
2350 This fix has been applied to the mysql and pgsql lookups by writing a
2351 separate quoting function and calling it from the lookup code. The fix
2352 should probably also be applied to nisplus, ibase and oracle lookups, but
2353 since I cannot test any of those, I have not disturbed their existing code.
2355 2. A hit in the callout cache for a specific address caused a log line with no
2356 reason for rejecting RCPT. Now it says "Previous (cached) callout
2357 verification failure".
2359 3. There was an off-by-one bug in the queryprogram router. An over-long
2360 return line was truncated at 256 instead of 255 characters, thereby
2361 overflowing its buffer with the terminating zero. As well as fixing this, I
2362 have increased the buffer size to 1024 (and made a note to document this).
2364 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2365 when Exim is waiting for an SMTP response from a remote server, Exim
2366 restarts its select() call on the socket, thereby resetting its timeout.
2367 This is not a problem when such interrupts are rare. Somebody set up a cron
2368 job to run exiwhat every 2 minutes, which is less than the normal select()
2369 timeout (5 or 10 minutes). This meant that the select() timeout never
2370 kicked in because it was always reset. I have fixed this by comparing the
2371 time when an interrupt arrives with the time at the start of the first call
2372 to select(). If more time than the timeout has elapsed, the interrupt is
2373 treated as a timeout.
2375 5. Some internal re-factoring in preparation for the addition of Sieve
2376 extensions (by MH). In particular, the "personal" test is moved to a
2377 separate function, and given an option for scanning Cc: and Bcc: (which is
2378 not set for Exim filters).
2380 6. When Exim created an email address using the login of the caller as the
2381 local part (e.g. when creating a From: or Sender: header line), it was not
2382 quoting the local part when it contained special characters such as @.
2384 7. Installed new OpenBSD configuration files.
2386 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2387 try to make them clearer.
2389 9. Callout options, other than the timeout value, were being ignored when
2390 verifying sender addresses in header lines. For example, when using
2392 verify = header_sender/callout=no_cache
2394 the cache was (incorrectly) being used.
2396 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2397 only the Cygwin environment.
2399 11. When return_path_on_delivery was set as a log selector, if different remote
2400 addresses in the same message used different return paths and parallel
2401 remote delivery occurred, the wrong values would sometimes be logged.
2402 (Whenever a remote delivery process finished, the return path value from
2403 the most recently started remote delivery process was logged.)
2405 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2406 lines when AUTH and/or TLS are in use. This is the "received protocol"
2407 field. Exim used to use "asmtp" for authenticated SMTP, without any
2408 indication (in the protocol name) for TLS use. Now it follows the RFC and
2409 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2410 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2411 names appear in log lines as well as in Received: header lines.
2413 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2414 extensions, and comparison tests, and to fix some bugs.
2416 14. Changes to the "personal" filter test:
2418 (1) The test was buggy in that it was just doing the equivalent of
2419 "contains" tests on header lines. For example, if a user's address was
2420 anne@some.where, the "personal" test would incorrectly be true for
2422 To: susanne@some.where
2424 This test is now done by extracting each address from the header in turn,
2425 and checking the entire address. Other tests that are part of "personal"
2426 are now done using regular expressions (for example, to check local parts
2427 of addresses in From: header lines).
2429 (2) The list of non-personal local parts in From: addresses has been
2430 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2431 taken from the Sieve specification recommendations.
2433 (3) If the message contains any header line starting with "List-" it is
2434 treated as non-personal.
2436 (4) The test for "circular" in the Subject: header line has been removed
2437 because it now seems ill-conceived.
2439 15. Minor typos in src/EDITME comments corrected.
2441 16. Installed latest exipick from John Jetmore.
2443 17. If headers_add on a router specified a text string that was too long for
2444 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2445 of string_sprintf() is now avoided.
2447 18. $message_body_size was not set (it was always zero) when running the DATA
2448 ACL and the local_scan() function.
2450 19. For the "mail" command in an Exim filter, no default was being set for
2451 the once_repeat time, causing a random time value to be used if "once" was
2452 specified. (If the value happened to be <= 0, no repeat happened.) The
2453 default is now 0s, meaning "never repeat". The "vacation" command was OK
2454 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2455 (I found it when inspecting the code).
2457 20. There is now an overall timeout for performing a callout verification. It
2458 defaults to 4 times the callout timeout, which applies to individual SMTP
2459 commands during the callout. The overall timeout applies when there is more
2460 than one host that can be tried. The timeout is checked before trying the
2461 next host. This prevents very long delays if there are a large number of
2462 hosts and all are timing out (e.g. when the network connections are timing
2463 out). The value of the overall timeout can be changed by specifying an
2464 additional sub-option for "callout", called "maxwait". For example:
2466 verify = sender/callout=5s,maxwait=20s
2468 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2469 to the end before writing, but this should make it even safer).
2471 22. Exim was forgetting that it had advertised PIPELINING for the second and
2472 subsequent messages on an SMTP connection. It was also not resetting its
2473 memory on STARTTLS and an internal HELO.
2475 23. When Exim logs an SMTP synchronization error within a session, it now
2476 records whether PIPELINING has been advertised or not.
2478 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2479 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2480 rather than long int.
2482 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2484 26. Added the never_mail option to autoreply.
2490 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2491 crash if the getsockname() call failed; this can happen if a connection is
2492 closed very soon after it is established. The problem was simply in the
2493 order in which certain operations were done, causing Exim to try to write
2494 to the SMTP stream before it had set up the file descriptor. The bug has
2495 been fixed by making things happen in the correct order.
2501 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2502 before the connection was closed, thus losing the rejection response.
2504 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2505 some early Solaris releases, but causes trouble in current releases where
2506 socklen_t is defined.
2508 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2511 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2514 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2515 that releasing the top of it at the end releases what was used for sub-
2516 expansions (unless the block got too big). However, discard this block if
2517 the first thing is a variable or header, so that we can use its block when
2518 it is dynamic (useful for very large $message_headers, for example).
2520 6. Lookups now cache *every* query, not just the most recent. A new, separate
2521 store pool is used for this. It can be recovered when all lookup caches are
2522 flushed. Lookups now release memory at the end of their result strings.
2523 This has involved some general refactoring of the lookup sources.
2525 7. Some code has been added to the store_xxx() functions to reduce the amount
2526 of flapping under certain conditions.
2528 8. log_incoming_interface used to affect only the <= reception log lines. Now
2529 it causes the local interface and port to be added to several more SMTP log
2530 lines, for example "SMTP connection from", and rejection lines.
2532 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2534 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2536 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2537 could be overwritten at the end of the current message (or the start of a
2538 new message if it was set in a HELO ACL). The value is now preserved for
2539 the duration of the SMTP connection.
2541 12. If a transport had a headers_rewrite setting, and a matching header line
2542 contained an unqualified address, that address was qualified, even if it
2543 did not match any rewriting rules. The underlying bug was that the values
2544 of the flags that permit the existence of unqualified sender and recipient
2545 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2546 non-local messages, and by -bnq for local messages) were not being
2547 preserved with the message after it was received.
2549 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2550 "next input=" as part of the text comprising the host identity instead of
2551 the correct text. The code was using the same buffer for two different
2552 strings. However, depending on which order the printing function evaluated
2553 its arguments, the bug did not always show up. Under Linux, for example, my
2554 test suite worked just fine.
2556 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2557 doesn't work with some older versions of Perl. It has been changed to "my",
2558 which in any case is probably the better facility to use.
2560 15. A really picky compiler found some instances of statements for creating
2561 error messages that either had too many or two few arguments for the format
2564 16. The size of the buffer for calls to the DNS resolver has been increased
2565 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2566 for addresses that have a lot of PTR records. This alleviates a problem; it
2567 does not fully solve it.
2569 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2570 buffer now truncates the list and logs the incident, which is the same
2571 action as happens when Exim is looking up a host name and its aliases.
2572 Previously in this situation something unpredictable would happen;
2573 sometimes it was "internal error: store_reset failed".
2575 18. If a server dropped the connection unexpectedly when an Exim client was
2576 using GnuTLS and trying to read a response, the client delivery process
2577 crashed while trying to generate an error log message.
2579 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2580 string, for example:
2582 warn message = H1: something\nH2: something
2584 the text was added as a single header line from Exim's point of view
2585 though it ended up OK in the delivered message. However, searching for the
2586 second and subsequent header lines using $h_h2: did not work. This has been
2587 fixed. Similarly, if a system filter added multiple headers in this way,
2588 the routers could not see them.
2590 20. Expanded the error message when iplsearch is called with an invalid key to
2591 suggest using net-iplsearch in a host list.
2593 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2594 ACLs are no longer actually imposed (and a message to that effect is
2597 22. If a "gecos" field in a passwd entry contained escaped characters, in
2598 particular, if it contained a \" sequence, Exim got it wrong when building
2599 a From: or a Sender: header from that name. A second bug also caused
2600 incorrect handling when an unquoted " was present following a character
2601 that needed quoting.
2603 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2604 was not being matched caselessly.
2606 24. Arranged for all hyphens in the exim.8 source to be escaped with
2609 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2610 properly. Recipient callout cache records were still being keyed to include
2611 the sender, even when use_sender was set false. This led to far more
2612 callouts that were necessary. The sender is no longer included in the key
2613 when use_sender is false.
2615 26. Added "control = submission" modifier to ACLs.
2617 27. Added the ${base62d: operator to decode base 62 numbers.
2619 28. dnsdb lookups can now access SRV records.
2621 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2622 the configuration file.
2624 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2625 (-v) mode. This makes the output for a verbose queue run more intelligible.
2627 31. Added a use_postmaster feature to recipient callouts.
2629 32. Added the $body_zerocount variable, containing the number of binary zero
2630 bytes in the message body.
2632 33. The time of last modification of the "new" subdirectory is now used as the
2633 "mailbox time last read" when there is a quota error for a maildir
2636 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2638 35. Added +ignore_unknown as a special item in host lists.
2640 36. Code for decoding IPv6 addresses in host lists is now included, even if
2641 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2642 address was recognized as an IP address, but was then not correctly decoded
2643 into binary, causing unexpected and incorrect effects when compared with
2650 1. Very minor rewording of debugging text in manualroute to say "list of
2651 hosts" instead of "hostlist".
2653 2. If verify=header_syntax was set, and a header line with an unqualified
2654 address (no domain) and a large number of spaces between the end of the
2655 name and the colon was received, the reception process suffered a buffer
2656 overflow, and (when I tested it) crashed. This was caused by some obsolete
2657 code that should have been removed. The fix is to remove it!
2659 3. When running in the test harness, delay a bit after writing a bounce
2660 message to get a bit more predictability in the log output.
2662 4. Added a call to search_tidyup() just before forking a reception process. In
2663 theory, someone could use a lookup in the expansion of smtp_accept_max_
2664 per_host which, without the tidyup, could leave open a database connection.
2666 5. Added the variables $recipient_data and $sender_data which get set from a
2667 lookup success in an ACL "recipients" or "senders" condition, or a router
2668 "senders" option, similar to $domain_data and $local_part_data.
2670 6. Moved the writing of debug_print from before to after the "senders" test
2673 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2674 problems for message scanning, either using a data ACL, or using
2675 local_scan() because the Received: header was not generated till after they
2676 were called (in order to set the time as the time of reception completion).
2677 I have revised the way this works. The header is now generated after the
2678 body is received, but before the ACL or local_scan() are called. After they
2679 are run, the timestamp in the header is updated.
2685 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2686 before starting a queue runner without re-exec. This happened only when
2687 deliver_drop_privilege was set or when the Exim user was set to root. The
2688 effect of the bug was that timeouts during subsequent deliveries caused
2689 crashes instead of being properly handled. The handler is now left at its
2690 default (and expected) setting.
2692 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2693 message, again when deliver_drop_privilege is set or Exim is run as root.
2694 The bug described in (1) was not present in this case, but the tidying up
2695 of the other signals was missing. I have made the two cases consistent.
2697 3. The ignore_target_hosts setting on a manualroute router was being ignored
2698 for hosts that were looked up using the /MX notation.
2700 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2703 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2704 operated on the sender address. After changing the $sender_address to <>
2705 for the sender address verify, Exim was re-instated it as the original
2706 (before rewriting) address, but remembering that it had rewritten it, so it
2707 wasn't rewriting it again. This bug also had the effect of breaking the
2708 sender address verification caching when the sender address was rewritten.
2710 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2711 This has been changed so that if the ip literal address matches
2712 ignore_target_hosts, the router declines.
2714 7. Added expansion conditions match_domain, match_address, and match_local_
2715 part (NOT match_host).
2717 8. The placeholder for the Received: header didn't have a length field set.
2719 9. Added code to Exim itself and to exim_lock to test for a specific race
2720 condition that could lead to file corruption when using MBX delivery. The
2721 issue is with the lockfile that is created in /tmp. If this file is removed
2722 after a process has opened it but before that process has acquired a lock,
2723 there is the potential for a second process to recreate the file and also
2724 acquire a lock. This could lead to two Exim processes writing to the file
2725 at the same time. The added code performs the same test as UW imapd; it
2726 checks after acquiring the lock that its file descriptor still refers to
2727 the same named file.
2729 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2730 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2733 11. Added the smtp_active_hostname option. If used, this will typically be made
2734 to depend on the incoming interface address. Because $interface_address is
2735 not set up until the daemon has forked a reception process, error responses
2736 that can happen earlier (such as "too many connections") no longer contain
2739 12. If an expansion in a condition on a "warn" statement fails because a lookup
2740 defers, the "warn" statement is abandoned, and the next ACL statement is
2741 processed. Previously this caused the whole ACL to be aborted.
2743 13. Added the iplsearch lookup type.
2745 14. Added ident_timeout as a log selector.
2747 15. Added tls_certificate_verified as a log selector.
2749 16. Added a global option tls_require_ciphers (compare the smtp transport
2750 option of the same name). This controls incoming TLS connections.
2752 17. I finally figured out how to make tls_require_ciphers do a similar thing
2753 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2754 before starting the TLS session.
2756 18. Tabs are now shown as \t in -bP output.
2758 19. If the log selector return_path_on_delivery was set, Exim crashed when
2759 bouncing a message because it had too many Received: header lines.
2761 20. If two routers both had headers_remove settings, and the first one included
2762 a superfluous trailing colon, the final name in the first list and the
2763 first name in the second list were incorrectly joined into one item (with a
2764 colon in the middle).
2770 1. Added -C and -D options to the exinext utility, mainly to make it easier
2771 to include in the automated testing, but these could be helpful when
2772 multiple configurations are in use.
2774 2. The exinext utility was not formatting the output nicely when there was
2775 an alternate port involved in the retry record key, nor when there was a
2776 message id as well (for retries that were specific to a specific message
2777 and a specific host). It was also confused by IPv6 addresses, because of
2778 the additional colons they contain. I have fixed the IPv4 problem, and
2779 patched it up to do a reasonable job for IPv6.
2781 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2782 delivery, the log line now contains "pipelined" if PIPELINING was used.
2784 4. An SMTP transport process used to panic and die if the bind() call to set
2785 an explicit outgoing interface failed. This has been changed; it is now
2786 treated in the same way as a connect() failure.
2788 5. A reference to $sender_host_name in the part of a conditional expansion
2789 that was being skipped was still causing a DNS lookup. This no longer
2792 6. The def: expansion condition was not recognizing references to header lines
2793 that used bh_ and bheader_.
2795 7. Added the _cache feature to named lists.
2797 8. The code for checking quota_filecount in the appendfile transport was
2798 allowing one more file than it should have been.
2800 9. For compatibility with Sendmail, the command line option
2808 and sets the incoming protocol and host name (for trusted callers). The
2809 host name and its colon can be omitted when only the protocol is to be set.
2810 Note the Exim already has two private options, -pd and -ps, that refer to
2811 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2812 "s", but I don't think that's a major issue.
2814 10. A number of refactoring changes to the code, none of which should affect
2817 (a) The number of logging options was getting close to filling up the
2818 32-bit word that was used as a bit map. I have split them into two classes:
2819 those that are passed in the argument to log_write(), and those that are
2820 only ever tested independently outside of that function. These are now in
2821 separate 32-bit words, so there is plenty of room for expansion again.
2822 There is no change in the user interface or the logging behaviour.
2824 (b) When building, for example, log lines, the code previously used a
2825 macro that called string_cat() twice, in order to add two strings. This is
2826 not really sufficiently general. Furthermore, there was one instance where
2827 it was actually wrong because one of the argument was used twice, and in
2828 one call a function was used. (As it happened, calling the function twice
2829 did not affect the overall behaviour.) The macro has been replaced by a
2830 function that can join an arbitrary number of extra strings onto a growing
2833 (c) The code for expansion conditions now uses a table and a binary chop
2834 instead of a serial search (which was left over from when there were very
2835 few conditions). Also, it now recognizes conditions like "pam" even when
2836 the relevant support is not compiled in: a suitably worded error message is
2837 given if an attempt is made to use such a condition.
2839 11. Added ${time_interval:xxxxx}.
2841 12. A bug was causing one of the ddress fields not to be passed back correctly
2842 from remote delivery subprocesses. The field in question was not being
2843 subsequently used, so this caused to problems in practice.
2845 13. Added new log selectors queue_time and deliver_time.
2847 14. Might have fixed a bug in maildirsizefile handling that threw up
2848 "unexpected character" debug warnings, and recalculated the data
2849 unnecessarily. In any case, I expanded the warning message to give more
2852 15. Added the message "Restricted characters in address" to the statements in
2853 the default ACL that block characters like @ and % in local parts.
2855 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2856 Three changes have been made:
2858 (a) There was a serious bug; a negative response to MAIL caused the whole
2859 recipient domain to be cached as invalid, thereby blocking all messages
2860 to all local parts at the same domain, from all senders. This bug has
2861 been fixed. The domain is no longer cached after a negative response to
2862 MAIL if the sender used is not empty.
2864 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2867 (c) A new callout option, "use_sender" has been added for people who want
2868 the modified behaviour.
2874 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2877 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2878 indeed breaks things for older releases.
2880 3. Added additional logging to the case where there is a problem reading data
2881 from a filter that is running in a subprocess using a pipe, in order to
2882 try to track down a specific problem.
2884 4. Testing facility fudge: when running in the test harness and attempting
2885 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2886 getting "No route to host". Convert this to a timeout.
2888 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2891 6. Some OS don't have socklen_t but use size_t instead. This affects the
2892 fifth argument of getsockopt() amongst other things. This is now
2893 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2894 can be set for individual OS. I have set it for SunOS5, OSF1, and
2895 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2896 some earlier ones do not.
2898 7. Change 4.30/15 was not doing the test caselessly.
2900 8. The standard form for an IPv6 address literal was being rejected by address
2901 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2902 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2903 this, as well as the form without the "IPv6" on the front (but only when
2904 address literals are enabled, of course).
2906 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2908 10. Exim crashed if a message with an empty sender address specified by -f
2909 encountered a router with an errors_to setting. This could be provoked only
2910 by a command such as
2914 where an empty string was supplied; "<>" did not hit this bug.
2916 11. Installed PCRE release 4.5.
2918 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2919 remained set. It is now erased.
2921 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2922 times from message ids (which are base 36 rather than the normal 62).
2924 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2925 were being counted as actual protocol errors, and logged if the log
2926 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2927 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2928 and DATA following a set of rejected RCPTs do not count as protocol errors.
2929 In other words, Exim assumes they were pipelined, though this may not
2930 actually be the case. Of course, in all cases the client gets an
2931 appropriate error code.
2933 15. If a lookup fails in an ACL condition, a message about the failure may
2934 be available; it is used if testing the ACL cannot continue, because most
2935 such messages specify what the cause of the deferral is. However, some
2936 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2937 that caused an old message to be retained and used if a later statement
2938 caused a defer, replacing the real cause of the deferral.
2940 16. If an IP address had so many PTR records that the DNS lookup buffer
2941 was not large enough to hold them, Exim could crash while trying to process
2942 the truncated data. It now detects and logs this case.
2944 17. Further to 4.21/58, another change has been made: if (and only if) the
2945 first line of a message (the first header line) ends with CRLF, a bare LF
2946 in a subsequent header line has a space inserted after it, so as not to
2947 terminate the header.
2949 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2950 unnecessarily, used atoi() instead of strtol(), and didn't check the
2951 termination when getting file sizes from file names by regex.
2953 19. Completely re-implemented the support for maildirsize files, in the light
2954 of a number of problems with the previous contributed implementation
2955 (4.30/29). In particular:
2957 . If the quota is zero, the maildirsize file is maintained, but no quota is
2960 . If the maildir directory does not exist, it is created before any attempt
2961 to write a maildirsize file.
2963 . The quota value in the file is just a cache; if the quota is changed in
2964 the transport, the new value overrides.
2966 . A regular expression is available for excluding directories from the
2969 20. The autoreply transport checks the characters in options that define the
2970 message's headers; it allows continued headers, but it was checking with
2971 isspace() after an embedded newline instead of explicitly looking for a
2974 21. If all the "regular" hosts to which an address was routed had passed their
2975 expiry times, and had not reached their retry times, the address was
2976 bounced, even if fallback hosts were defined. Now Exim should go on to try
2979 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2980 equivalent code in the SMTP transport. Some hosts send humungous responses
2981 to HELO/EHLO, more than 1024 it seems.
2983 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2984 gives compiler warnings in some environments. I've now done it "properly",
2987 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2988 (because of problems with the built-in one) was declared to return uschar *
2989 instead of char *, causing compiler failure.
2991 25. Fixed a file descriptor leak when processing alias/forward files.
2993 26. Fixed a minor format string issue in dbfn.c.
2995 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2997 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2998 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3001 29. When a sender address is verified, it is cached, to save repeating the test
3002 when there is more than one recipient in a message. However, when the
3003 verification involves a callout, it is possible for different callout
3004 options to be set for different recipients. It is too complicated to keep
3005 track of this in the cache, so now Exim always runs a verification when a
3006 callout is required, relying on the callout cache for the optimization.
3007 The overhead is duplication of the address routing, but this should not be
3010 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3011 to be verified with callout=postmaster, and the main callout worked but the
3012 postmaster check failed, the verification correctly failed. However, if a
3013 subsequent RCPT command asked for sender verification *without* the
3014 postmaster check, incorrect caching caused this verification also to fail,
3017 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3018 it was not caching the DNS options (qualify_single, search_parents) that
3019 were used when the lookup failed. A subsequent lookup with different
3020 options therefore always gave the same answer, though there were cases
3021 where it should not have. (Example: a "domains = !$mx_any" option on a
3022 dnslookup router: the "domains" option is always processed without any
3023 widening, but the router might have qualify_single set.) Now Exim uses the
3024 cached value only when the same options are set.
3026 32. Added John Jetmore's "exipick" utility to the distribution.
3028 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3029 than a timeout (e.g. a certificate is required, and is not provided), an
3030 Exim server now closes the connection immediately. Previously it waited for
3031 the client to close - but if the client is SSL, it seems that they each
3032 wait for each other, leading to a delay before one of them times out.
3034 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3035 maintained 0.8.x compatibility because I don't think many are using it, and
3036 it is clearly obsolete.
3038 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3041 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3042 client certificate was expired. A simple patch fixes this, though I don't
3043 understand the full logic of why the verify callback is called multiple
3046 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3047 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3048 which causes problems with some clients (such as the Certicom SSL Plus
3049 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3050 disables the coutermeasure allowing Eudora to connect."
3052 38. Exim was not checking that a write() to a log file succeeded. This could
3053 lead to Bad Things if a log got too big, in particular if it hit a file
3054 size limit. Exim now panics and dies if it cannot write to a log file, just
3055 as it does if it cannot open a log file.
3057 39. Modified OS/Makefile-Linux so that it now contains
3059 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3061 The two -D definitions ensure that Exim is compiled with large file
3062 support, which makes it possible to handle log files that are bigger than
3065 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3066 instance) a domain was checked against a named list that involved a lookup,
3067 causing $domain_data to be set, then another domain was checked against the
3068 same list, then the first domain was re-checked, the value of $domain_data
3069 after the final check could be wrong. In particular, if the second check
3070 failed, it could be set empty. This bug probably also applied to
3073 41. The strip_trailing_dot option was not being applied to the address given
3074 with the -f command-line option.
3076 42. The code for reading a message's header from the spool was incrementing
3077 $received_count, but never initializing it. This meant that the value was
3078 incorrect (doubled) while delivering a message in the same process in which
3079 it was received. In the most common configuration of Exim, this never
3080 happens - a fresh exec is done - but it can happen when
3081 deliver_drop_privilege is set.
3083 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3084 it now includes up to 150 characters of the unexpected data in the log
3087 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3088 and building data strings. The size of both of these buffers was 10 000
3089 bytes - far larger than anybody would *ever* want, thought I. Needless to
3090 say, somebody hit the limit. I have increased the maximum line length to
3091 20 000 and the maximum data length of concatenated lines to 100 000. I have
3092 also fixed two bugs, because there was no checking on these buffers. Tsk,
3093 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3094 buffer is too small.
3096 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3097 lsearch lookups. Now it does.
3099 46. When parsing a route_list item in a manualroute router, a fixed-length
3100 buffer was used for the list of hosts. I made this 1024 bytes long,
3101 thinking that nobody would ever have a list of hosts that long. Wrong.
3102 Somebody had a whole pile of complicated expansion conditions, and the
3103 string was silently truncated, leading to an expansion error. It turns out
3104 that it is easier to change to an unlimited length (owing to other changes
3105 that have happened since this code was originally written) than to build
3106 structure for giving a limitation error. The length of the item that
3107 expands into the list of hosts is now unlimited.
3109 47. The lsearch lookup could not handle data where the length of text line was
3110 more than 4095 characters. Such lines were truncated, leading to shortened
3111 data being returned. It should now handle lines of any length.
3113 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3114 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3117 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3118 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3119 no longer happens in comment lines. A list of replacements is now placed
3120 at the head of all of the source files, except those whose only change is
3121 to replace PERL_COMMAND in the very first #! line.
3123 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3124 messages on the queue, with a bottom-up merge sort, using code contributed
3125 by Michael Haardt. This should make operations like -bp somewhat faster on
3126 large queues. It won't affect queue runners, except when queue_run_in_order
3129 51. Installed eximstats 1.31 in the distribution.
3131 52. Added support for SRV lookups to the dnslookup router.
3133 53. If an ACL referred to $message_body or $message_body_end, the value was not
3134 reset for any messages that followed in the same SMTP session.
3136 54. The store-handling optimization for building very long strings was not
3137 differentiating between the different store pools. I don't think this
3138 actually made any difference in practice, but I've tidied it.
3140 55. While running the routers to verify a sender address, $sender_address
3141 was still set to the sender address. This is wrong, because when routing to
3142 send a bounce to the sender, it would be empty. Therefore, I have changed
3143 it so that, while verifying a sender address, $sender_address is set to <>.
3144 (There is no change to what happens when verifying a recipient address.)
3146 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3147 target A or AAAA records (if not already returned) without resetting the
3148 qualify_single or search_parents options of the DNS resolver. These are
3149 inappropriate in this case because the targets of MX and SRV records must
3150 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3151 target that, when qualified, matched something in the local domain. These
3152 two options are now turned off when doing these lookups.
3154 57. It seems that at least some releases of Reiserfs (which does not have the
3155 concept of a fixed number of inodes) returns zero and not -1 for the
3156 number of available inodes. This interacted badly with check_spool_inodes,
3157 which assumed that -1 was the "no such thing" setting. What I have done is
3158 to check that the total number of inodes is greater than zero before doing
3159 the test of how many are available.
3161 58. When a "warn" ACL statement has a log_message modifier, the message is
3162 remembered, and not repeated. This is to avoid a lot of repetition when a
3163 message has many recipients that cause the same warning to be written.
3164 Howewer, Exim was preserving the list of already written lines for an
3165 entire SMTP session, which doesn't seem right. The memory is now reset if a
3166 new message is started.
3168 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3169 debugging output unless log_rewrite was also set.
3171 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3172 of (int)(handle) when we know that handle contains (void *)(-1).
3174 61. The Exim daemon panic-logs an error return when it closes the incoming
3175 connection. However "connection reset by peer" seems to be common, and
3176 isn't really an error worthy of noting specially, so that particular error
3179 62. When Exim is trying to find all the local interfaces, it used to panic and
3180 die if the ioctl to get the interface flags failed. However, it seems that
3181 on at least one OS (Solaris 9) it is possible to have an interface that is
3182 included in the list of interfaces, but for which you get a failure error
3183 for this call. This happens when the interface is not "plumbed" into a
3184 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3185 failure of the "get flags" call assumes that the interface is down.
3187 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3188 makes life easier for people who are doing arithmetic on fields extracted
3189 from dates, where you often get leading zeros that should not be
3190 interpreted as octal.
3192 64. Added qualify_domain to the redirect router, to override the global
3195 65. If a pathologically long header line contained very many addresses (the
3196 report of this problem mentioned 10 000) and each of them was rewritten,
3197 Exim could use up a very large amount of memory. (It kept on making new
3198 copies of the header line as it rewrote, and never released the old ones.)
3199 At the expense of a bit more processing, the header rewriting function has
3200 been changed so that it no longer eats memory in this way.
3202 66. The generation of the Received: header has been moved from the time that a
3203 message starts to be received, to the time that it finishes. The timestamp
3204 in the Received: header should now be very close to that of the <= log
3205 line. There are two side-effects of this change:
3207 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3208 logged header lines no longer include the local Received: line, because
3209 it has not yet been created. The same applies to a copy of the message
3210 that is returned to a non-SMTP sender when a message is rejected.
3212 (b) When a filter file is tested using -bf, no additional Received: header
3213 is added to the test message. After some thought, I decided that this
3216 This change does not affect the value of $received_for. It is still set
3217 after address rewriting, but before local_scan() is called.
3219 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3221 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3222 gave an unhelpful panic error message, and a defer error. I have managed to
3223 change this behaviour so that it now rejects any supplied certificate,
3224 which seems right, as the list of acceptable certificates is empty.
3226 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3227 gave an unhelpful defer error. I have not managed to make this reject any
3228 supplied certificates, but the error message it gives is "no certificate
3229 supplied", which is not helpful.
3231 70. exigrep's output now also includes lines that are not associated with any
3232 message, but which match the given pattern. Implemented by a patch from
3233 Martin Sluka, which also tidied up the Perl a bit.
3235 71. Recipient callout verification, like sender verification, was using <> in
3236 the MAIL FROM command. This isn't really the right thing, since the actual
3237 sender may affect whether the remote host accepts the recipient or not. I
3238 have changed it to use the actual sender in the callout; this means that
3239 the cache record is now keyed on a recipient/sender pair, not just the
3240 recipient address. There doesn't seem to be a real danger of callout loops,
3241 since a callout by the remote host to check the sender would use <>.
3242 [SEE ABOVE: changed after hitting problems.]
3244 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3245 temporary errors. However, in the case of such a code being given after
3246 the end of a data transmission (i.e. after ".") Exim was failing to write
3247 a retry record for the message. (Yes, there was some broken host that was
3248 actually sending 8xx at this point.)
3250 73. An unknown lookup type in a host list could cause Exim to panic-die when
3251 the list was checked. (An example that provoked this was putting <; in the
3252 middle of a list instead of at the start.) If this happened during a DATA
3253 ACL check, a -D file could be left lying around. This kind of configuration
3254 error no longer causes Exim to die; instead it causes a defer errror. The
3255 incident is still logged to the main and panic logs.
3257 74. Buglet left over from Exim 3 conversion. The message "too many messages
3258 in one connection" was written to the rejectlog but not the mainlog, except
3259 when address rewriting (yes!) was being logged.
3261 75. Added write_rejectlog option.
3263 76. When a system filter was run not as root (that is, when system_filter_user
3264 was set), the values of the $n variables were not being returned to the
3265 main process; thus, they were not subsequently available in the $sn
3268 77. Added +return_path_on_delivery log selector.
3270 78. A connection timeout was being treated differently from recipients deferred
3271 when testing hosts_max_try with a message that was older than the host's
3272 retry timeout. (The host should not be counted, thus allowing all hosts to
3273 be tried at least once before bouncing.) This may have been the cause of an
3274 occasionally reported bug whereby a message would remain on the queue
3275 longer than the retry timeout, but would be bounced if a delivery was
3276 forced. I say "may" because I never totally pinned down the problem;
3277 setting up timeout/retry tests is difficult. See also the next item.
3279 79. The ultimate address timeout was not being applied to errors that involved
3280 a combination of host plus message (for example, a timeout on a MAIL
3281 command). When an address resolved to a number of possible hosts, and they
3282 were not all tried for each delivery (e.g. because of hosts_max_try), a
3283 message could remain on the queue longer than the retry timeout.
3285 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3288 81. Fixed an obscure SMTP outgoing bug which required at least the following
3289 conditions: (a) there was another message waiting for the same server;
3290 (b) the server returned 5xx to all RCPT commands in the first message so
3291 that the message was not completed; (c) the server dropped the connection
3292 or gave a negative response to the RSET that Exim sends to abort the
3293 transaction. The observed case was a dropped connection after DATA that had
3294 been sent in pipelining mode. That is, the server had advertised PIPELINING
3295 but was not implementing it correctly. The effect of the bug was incorrect
3296 behaviour, such as trying another host, and this could lead to a crash.
3302 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3303 and daemon.c were passed as pointers to ints; they should have been
3304 pointers to socklen_t variables (which are typically unsigned ints).
3306 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3309 3. Fixed a really odd bug that affected only the testing scheme; patching a
3310 certain fixed string in the binary changed the value of another string that
3311 happened to be identical to the end of the original first string.
3313 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3314 name", it returns that address as the IP address. On some operating
3315 systems (e.g. Solaris), it also passes back the IP address string as the
3316 "host name". However, on others (e.g. Linux), it passes back an empty
3317 string. Exim wasn't checking for this, and was changing the host name to an
3318 empty string, assuming it had been canonicized.
3320 5. Although rare, it is permitted to have more than one PTR record for a given
3321 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3322 all the names associated with an address, because they do in Solaris.
3323 However, it seems that they do not in Linux for data that comes from the
3324 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3325 I found this out when I moved to a new Linux workstation and tried to run
3326 the Exim test suite.
3328 To get round this problem I have changed the code so that it now does its
3329 own call to the DNS to look up PTR records when searching for a host name.
3330 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3331 addresses that are only in /etc/hosts are still found.
3333 This behaviour is, however, controlled by an option called host_lookup_
3334 order, which defaults to "bydns:byaddr". If people want to use the other
3335 order, or indeed, just use one or the other means of lookup, they can
3336 specify it in this variable.
3338 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3339 some operating systems, this comes back from gethostbyaddr() as an empty
3340 string, and this is what Exim used to test for. However, it seems that in
3341 other systems, "." is yielded. Exim now tests for this case too.
3343 7. The values of check_spool_space and check_log_space are now held internally
3344 as a number of kilobytes instead of an absolute number of bytes. If a
3345 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3346 kilobyte. This means that much larger values can be stored.
3348 8. Exim monitor: an attempt to get the action menu when not actually pointing
3349 at a message produces an empty menu entitled "No message selected". This
3350 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3351 no entries in it ("Shell widget menu has zero width and/or height"). So I
3352 have added a single, blank menu entry in this case.
3354 9. Added ${quote_local_part.
3356 10. MIME decoding is now applied to the contents of Subject: header lines when
3359 11. Now that a reference to $sender_host_address automatically causes a reverse
3360 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3361 host lookup before query-style lookups in lists that might use this
3362 variable. This has therefore been abolished, and the "net-" prefix is no
3363 longer necessary for query-style lookups.
3365 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3366 have been a typo for LFLAGS, so it has been changed.
3368 13. The install script calls Exim with "-C /dev/null" in order to find the
3369 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3370 to be output. Howeve, since Exim outputs its version number before the
3371 error, it didn't break the script. It just looked ugly. I fixed this by
3372 always allowing "-C /dev/null" if the caller is root.
3374 14. Ignore overlarge ACL variable number when reading spool file - insurance
3375 against a later release with more variables having written the file.
3377 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3378 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3379 this, as well as the form without the "IPv6" on the front.
3381 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3382 OS/Makefile-Darwin file.
3384 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3385 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3387 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3388 variable. However, some OS use ints for inodes. I've added cast to long int
3389 to get rid of the compiler warning.
3391 19. I had forgotten to lock out "/../" in configuration file names when
3392 ALT_CONFIG_PREFIX was set.
3394 20. Routers used for verification do not need to specify transports. However,
3395 if such a router generated a host list, and callout was configured, Exim
3396 crashed, because it could not find a port number from the (non-existent)
3397 transport. It now assumes port 25 in this circumstance.
3399 21. Added the -t option to exigrep.
3401 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3402 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3403 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3404 them is set, LOOKUP_LSEARCH is forced.
3406 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3408 24. Added sender and host information to the "rejected by local_scan()" log
3409 line; previously there was no indication of these.
3411 25. Added .include_if_exists.
3413 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3414 Linux. It turns out that not all file systems support this. Apparently some
3415 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3416 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3417 not supported on the file descriptor, is now ignored when Exim is trying to
3418 sync a directory. This applies only to Linux.
3420 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3422 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3423 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3426 29. Added support for maildirsize files from supplied patch (modified a bit).
3428 30. The use of :fail: followed by an empty string could lead Exim to respond to
3429 sender verification failures with (e.g.):
3431 550 Verification failed for <xxx>
3432 550 Sender verify failed
3434 where the first response line was missing the '-' that indicates it is not
3435 the final line of the response.
3437 31. The loop for finding the name of the user that called Exim had a hardwired
3438 limit of 10; it now uses the value of finduser_retries, which is used for
3439 all other user lookups.
3441 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3444 33. Exim was neglecting to zero errno before one call of strtol() when
3445 expanding a string and expecting an integer value. On some systems this
3446 resulted in spurious "integer overflow" errors. Also, it was casting the
3447 result into an int without checking.
3449 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3450 did not work. The code looks as if it has *never* worked, though it appears
3451 to have been documented since at least releast 1.62. I have made it work.
3453 35. The "timeout_DNS" error in retry rules, also documented since at least
3454 1.62, also never worked. As it isn't clear exactly what this means, and
3455 clearly it isn't a major issue, I have abolished the feature by treating it
3456 as "timeout", and writing a warning to the main and panic logs.
3458 36. The display of retry rules for -brt wasn't always showing the error code
3461 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3462 timeout_connect_A, timeout_connect_MX.
3464 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3465 to the empty sender.
3467 39. The daemon was not analysing the content of -oX till after it had closed
3468 stderr and disconnected from the controlling terminal. This meant that any
3469 syntax errors were only noted on the panic log, and the return code from
3470 the command was 0. By re-arranging the code a little, I've made the
3471 decoding happen first, so such errors now appear on stderr, and the return
3472 code is 1. However, the actual setting up of the sockets still happens in
3473 the disconnected process, so errors there are still only recorded on the
3476 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3477 connections (as happens on some IP stacks) was logged at start up time as
3478 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3479 it from "IPv6 and IPv4", which means that two separate sockets are being
3482 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3483 says whether AF_INET or AF_INET6 was passed as an argument.
3485 42. Exiwhat output was messed up when time zones were included in log
3488 43. Exiwhat now gives more information about the daemon's listening ports,
3489 and whether -tls-on-connect was used.
3491 44. The "port" option of the smtp transport is now expanded.
3493 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3494 silently ignored. Now an error message is written to the main and panic
3497 46. There's a new ACL modifier called "logwrite" which writes to a log file
3498 as soon as it is encountered.
3500 47. Added $local_user_uid and $local_user_gid at routing time.
3502 48. Exim crashed when trying to verify a sender address that was being
3505 49. Exim was recognizing only a space character after ".include". It now also
3506 recognizes a tab character.
3508 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3509 extracting the relevant information from the specification. The man page no
3510 longer contains scrambled data for the -d option, and I've added a section
3511 at the front about calling Exim under different names.
3513 51. Added "extra_headers" argument to the "mail" command in filter files.
3515 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3518 53. Installed eximstats 1.29.
3520 54. Added transport_filter_timeout as a generic transport option.
3522 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3523 Cc: header lines. This was required by RFC 822, but it not required by RFC
3526 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3527 incoming messages that did not have them. Now it does so only if the
3528 message originates locally, that is, if there is no associated remote host
3529 address. When Resent- header lines are present, this applies to the Resent-
3530 lines rather than the non-Resent- lines.
3532 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3533 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3535 58. Messages for configuration errors now include the name of the main
3536 configuration file - useful now that there may be more than one file in a
3537 list (.included file names were always shown).
3539 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3540 for those rare installations that do not start the daemon as root or run it
3541 setuid root. I've cut out the call to initgroups() if the daemon is not
3544 60. The Exim user and group can now be bound into the binary as text strings
3545 that are looked up at the start of Exim's processing.
3547 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3549 62. Added $mailstore_basename variable.
3551 63. Installed patch to sieve.c from Michael Haardt.
3553 64. When Exim failed to open the panic log after failing to open the main log,
3554 the original message it was trying to log was written to stderr and debug
3555 output, but if they were not available (the usual case in production), it
3556 was lost. Now it is written to syslog before the two lines that record the
3557 failures to open the logs.
3559 65. Users' Exim filters run in subprocesses under the user's uid. It is
3560 possible for a "deliver" command or an alias in a "personal" command to
3561 provoke an address rewrite. If logging of address rewriting is configured,
3562 this fails because the process is not running as root or exim. There may be
3563 a better way of dealing with this, but for the moment (because 4.30 needs
3564 to be released), I have disabled address rewrite logging when running a
3565 filter in a non-root, non-exim process.
3571 1. The buildconfig auxiliary program wasn't quoting the value set for
3572 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3573 not defined. This bug was masked in 4.22 by the effect that was fixed in
3576 2. Some messages that were rejected after a message id was allocated were
3577 shown as "incomplete" by exigrep. It no longer does this for messages that
3578 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3580 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3581 have allow_domain_literals set, the ID did not get logged in the <= line.
3582 Domain literals are now always recognized in Message-ID: header lines.
3584 4. The first argument for a ${extract expansion item is the key name or field
3585 number. Leading and trailing spaces in this item were not being ignored,
3586 causing some misleading effects.
3588 5. When deliver_drop_privilege was set, single queue runner processes started
3589 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3590 same command in the process it spins off) were not dropping privilege.
3592 6. When the daemon running as "exim" started a queue runner, it always
3593 re-executed Exim in the spun-off process. This is a waste of effort when
3594 deliver_drop_privilege is set. The new process now just calls the
3595 queue-runner function directly.
3601 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3604 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3605 ignored. Though the use of -oP was forcing the writing of a pid file, it
3606 was always written to the default place.
3608 3. If the message "no IP address found for host xxxx" is generated during
3609 incoming verification, it is now followed by identification of the incoming
3610 connection (so you can more easily find what provoked it).
3612 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3614 5. Added some features to "harden" Exim a bit more against certain attacks:
3616 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3617 be put in Local/Makefile. This is like the never_users runtime option,
3618 but it cannot be overridden. The default setting is "root".
3620 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3621 prefix string with which any file named in a -C command line option
3624 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3625 is retained for -C and -D only if the caller of Exim is root. Without
3626 it, the exim user may also use -C and -D and retain privilege.
3628 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3629 command line option is disabled.
3631 6. Macro names set by the -D option must start with an upper case letter, just
3632 like macro names defined in the configuration file.
3634 7. Added "dereference=" facility to LDAP.
3636 8. Two instances of the typo "uknown" in the source files are fixed.
3638 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3639 the Configure-Makefile script screwed up while processing it.
3641 10. Incorporated PCRE 4.4.
3643 11. The SMTP synchronization check was not operating right at the start of an
3644 SMTP session. For example, it could not catch a HELO sent before the client
3645 waited for the greeting. There is now a check for outstanding input at the
3646 point when the greeting is written. Because of the duplex, asynchronous
3647 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3648 way, but not yet received, when the check is performed.
3650 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3651 on TCP/IP sockets, because this apparently causes some broken clients to
3654 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3655 unchanged) from the Cygwin maintainer.
3657 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3658 was supported instead of testing for mbx. Effectively a typo.
3660 15. The spa authenticator server code was not checking that the input it
3661 received was valid base64.
3663 16. The debug output line for the "set" modifier in ACLs was not showing the
3664 name of the variable that was being set.
3666 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3668 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3669 lookup on its own. Something else was needed to trigger the lookup. For
3670 example, a match in host_lookup or the need for a host name in a host list.
3671 Now, if $sender_host_name is referenced and the host name has not yet been
3672 looked up, a lookup is performed. If the lookup fails, the variable remains
3673 empty, and $host_lookup_failed is set to "1".
3675 19. Added "eqi" as a case-independent comparison operator.
3677 20. The saslauthd authentication condition could segfault if neither service
3678 nor realm was specified.
3680 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3681 error message that was logged was misleading, and incoming SMTP
3682 connections were dropped. The message is now more accurate, and temporary
3683 errors are given to SMTP connections.
3685 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3686 (except RSET) with a 421 error, until QUIT is received. However, it was
3687 failing to send a response to QUIT.
3689 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3690 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3691 ACL is now run after the helo_try_verify_hosts code.
3693 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3694 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3695 case-independent, so other case variants are also recognized). Apparently
3696 some systems use these upper case variants.
3698 25. If more than two messages were waiting for the same host, and a transport
3699 filter was specified for the transport, Exim sent two messages over the
3700 same TCP/IP connection, and then failed with "socket operation on non-
3701 socket" when it tried to send the third.
3703 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3705 27. The extern definition of crypt16() in expand.c was not being excluded when
3706 the OS had its own crypt16() function.
3708 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3709 as a preferred synonym for return_size_limit, both as an option and as an
3712 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3714 30. Changed the default configuration ACL to relax the local part checking rule
3715 for addresses that are not in any local domains. For these addresses,
3716 slashes and pipe symbols are allowed within local parts, but the sequence
3717 /../ is explicitly forbidden.
3719 31. SPA server authentication was not clearing the challenge buffer before
3722 32. log_message in a "warn" ACL statement was writing to the reject log as
3723 well as to the main log, which contradicts the documentation and doesn't
3724 seem right (because no rejection is happening). So I have stopped it.
3726 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3727 However, I am unable to do any testing of this.
3729 34. Fixed an infelicity in the appendfile transport. When checking directories
3730 for a mailbox, to see if any needed to be created, it was accidentally
3731 using path names with one or more superfluous leading slashes; tracing
3732 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3734 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3735 contents are added to the log line that is written for every discarded
3736 recipient. (Previously a log_message setting was ignored.)
3738 36. The ${quote: operator now quotes the string if it is empty.
3740 37. The install script runs exim in order to find its version number. If for
3741 some reason other than non-existence or emptiness, which it checks, it
3742 could not run './exim', it was installing it with an empty version number,
3743 i.e. as "exim-". This error state is now caught, and the installation is
3746 38. An argument was missing from the function that creates an error message
3747 when Exim fails to connect to the socket for saslauthd authentication.
3748 This could cause Exim to crash, or give a corrupted message.
3750 39. Added isip, isip4, and isip6 to ${if conditions.
3752 40. The ACL variables $acl_xx are now saved with the message, and can be
3753 accessed later in routers, transports, and filters.
3755 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3756 strings in the file are not string-expanded.
3758 42. If a MAIL command specified a SIZE value that was too large to fit into an
3759 int variable, the check against message_size_limit failed. Such values are
3760 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3761 day this will have to be increased, but I don't think I want to be around
3762 when emails are that large.
3769 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3770 iconv() is not standard in FreeBSD.
3772 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3773 IPv6 enabled. The observed symptom was a segmentation fault on return from
3774 the function os_common_find_running_interfaces() in src/os.c.
3776 3. In the check_special_case() function in daemon.c I had used "errno" as an
3777 argument name, which causes warnings on some systems. This was basically a
3778 typo, since it was named "eno" in the comments!
3780 4. The code that waits for the clock to tick (at a resolution of some fraction
3781 of a second) so as to ensure message-id uniqueness was always waiting for
3782 at least one whole tick, when it could have waited for less. [This is
3783 almost certainly not relevant at current processor speeds, where it is
3784 unlikely to ever wait at all. But we try to future-proof.]
3786 5. The function that sleeps for a time interval that includes fractions of a
3787 second contained a race. It did not block SIGALRM between setting the
3788 timer, and suspending (a couple of lines later). If the interval was short
3789 and the sigsuspend() was delayed until after it had expired, the suspension
3790 never ended. On busy systems this could lead to processes getting stuck for
3793 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3794 process, before it forks any delivery processes. The open lookup caching
3795 mechanism meant that the open file or database connection was passed into
3796 the delivery process. The problem was that delivery processes always tidy
3797 up cached lookup data. This could cause a problem for the next delivery
3798 process started by the queue runner, because the external queue runner
3799 process does not know about the closure. So the next delivery process
3800 still has data in the lookup cache. In the case of a file lookup, there was
3801 no problem because closing a file descriptor in a subprocess doesn't affect
3802 the parent. However, if the lookup was caching a connection to a database,
3803 the connection was closed, and the second delivery process was likely to
3804 see errors such as "PGSQL: query failed: server closed the connection
3805 unexpectedly". The problem has been fixed by closing all cached lookups
3806 in a queue runner before running a delivery process.
3808 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3809 seem to have the "const" qualifier which it has on other OS. I've
3812 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3813 *of the same type* (client or server) with the same public name that an
3814 error should be diagnosed.
3816 9. When Exim looked up a host name for an IP address, but failed to find the
3817 original IP address when looking up the host name (a safety check), it
3818 output the message "<ip address> does not match any IP for NULL", which was
3819 confusing, to say the least. The bug was that the host name should have
3820 appeared instead of "NULL".
3822 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3823 user that is built into the binary, and the -C or -D options is used, root
3824 privilege is dropped before the configuration file is read. In addition,
3825 logging is switched to stderr instead of the normal log files. If the
3826 configuration then re-defines the Exim user, the unprivileged environment
3827 is probably not what is expected, so Exim logs a panic warning message (but
3830 However, if deliver_drop_privilege is set, the unprivileged state may well
3831 be exactly what is intended, so the warning has been cut out in that case,
3832 and Exim is allowed to try to write to its normal log files.
3838 1. smtp_return_error_details was not giving details for temporary sender
3839 or receiver verification errors.
3841 2. Diagnose a configuration error if two authenticators have the same public
3844 3. Exim used not to create the message log file for a message until the first
3845 delivery attempt. This could be confusing when incoming messages were held
3846 for policy or load reasons. The message log file is now created at the time
3847 the message is received, and an initial "Received" line is written to it.
3849 4. The automatically generated man page for command line options had a minor
3850 bug that caused no ill effects; however, a more serious problem was that
3851 the procedure for building the man page automatically didn't always
3852 operate. Consequently, release 4.20 contains an out-of-date version. This
3853 shouldn't happen again.
3855 5. When building Exim with embedded Perl support, the script that builds the
3856 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3857 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3859 6. The freeze_tell option was not being used for messages that were frozen on
3860 arrival, either by an ACL or by local_scan().
3862 7. Added the smtp_incomplete_transaction log selector.
3864 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3865 was accepting AUTH without a new EHLO.
3867 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3868 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3871 10. Logging of TCP/IP connections (when configured) now happens in the main
3872 daemon process instead of the child process, so that the TCP/IP connection
3873 count is more accurate (but it can never be perfect).
3875 11. The use of "drop" in a nested ACL was not being handled correctly in the
3876 outer ACL. Now, if condition failure induced by the nested "drop" causes
3877 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3878 or "require"), the connection is dropped.
3880 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3881 that yield "discard" can now be used with an "accept" or a "discard" verb,
3882 but an error is generated for any others (because I can't see a useful way
3883 to define what should happen).
3885 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3886 are now processed in the same way as lines in the Exim configuration file.
3887 In particular, continuation lines are supported.
3889 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3891 15. Added -ti meaning -t -i.
3893 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3894 domains, and warn by logging if others are found.
3896 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3897 returned from GIFCONF when Exim is trying to find the list of interfaces on
3898 a host. The code in os.c has been modified to copy each ifreq to an aligned
3899 structure in all cases.
3901 Also, in some cases, the returned ifreq's were being copied to a 'struct
3902 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3903 means the last couple of bytes of an IPv6 address could be chopped if the
3904 ifreq contained only a normal sockaddr (14 bytes storage).
3906 18. Named domain lists were not supported in the hosts_treat_as_local option.
3907 An entry such as +xxxx was not recognized, and was treated as a literal
3910 19. Ensure that header lines added by a DATA ACL are included in the reject log
3911 if the ACL subsequently rejects the message.
3913 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3914 MD5 (which is deprecated).
3916 21. When testing a filter file using -bf, Exim was writing a message when it
3917 took the sender from a "From " line in the message, but it was not doing so
3918 when it took $return_path from a Return-Path: header line. It now does.
3920 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3921 with a valid header line field name (a series of printing characters
3922 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3924 23. Changed "disc" in the source to "disk" to conform to the documentation and
3925 the book and for uniformity.
3927 24. Ignore Sendmail's -Ooption=value command line item.
3929 25. When execve() failed while trying to run a command in a pipe transport,
3930 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3931 could be confused with a return value of 69 from the command itself. This
3932 has been changed to 127, the value the shell returns if it is asked to run
3933 a non-existent command. The wording for the related log line suggests a
3934 non-existent command as the problem.
3936 26. If received_header_text expands to an empty string, do not add a Received:
3937 header line to the message. (Well, it adds a token one on the spool, but
3938 marks it "old" so that it doesn't get used or transmitted.)
3940 27. Installed eximstats 1.28 (addition of -nt option).
3942 28. There was no check for failure on the call to getsockname() in the daemon
3943 code. This can fail if there is a shortage of resources on the system, with
3944 ENOMEM, for example. A temporary error is now given on failure.
3946 29. Contrary to the C standard, it seems that in some environments, the
3947 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3948 program. Exim now does this explicitly; it affects the formatting of
3949 timestamps using strftime().
3951 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3952 complaints. I've now initialized all the variables, to avoid this.
3954 32. Header lines added by a system filter were not being "seen" during
3955 transport-time rewrites.
3957 33. The info_callback() function passed to OpenSSL is set up with type void
3958 (*)(SSL *, int, int), as described somewhere. However, when calling the
3959 function (actually a macro) that sets it up, the type void(*)() is
3960 expected. I've put in a cast to prevent warnings from picky compilers.
3962 34. If a DNS black list lookup found a CNAME record, but there were no A
3963 records associated with the domain it pointed at, Exim crashed.
3965 35. If a DNS black list lookup returned more than one A record, Exim ignored
3966 all but the first. It now scans all returned addresses if a particular IP
3967 value is being sought. In this situation, the contents of the
3968 $dnslist_value variable are a list of all the addresses, separated by a
3971 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3972 to accept a host name and all its aliases if the forward lookup for any of
3973 them yielded the IP address of the incoming connection. Now it accepts only
3974 those names whose forward lookup yields the correct IP address. Any other
3975 names are discarded. This closes a loophole whereby a rogue DNS
3976 administrator could create reverse DNS records to break through a
3977 wildcarded host restriction in an ACL.
3979 37. If a user filter or a system filter that ran in a subprocess used any of
3980 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3981 the wrong values were passed to the pipe command ($thisaddress had the
3982 value of $0, $0 had the value of $1, etc). This bug was introduced by
3983 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3985 38. Improved the line breaking for long SMTP error messages from ACLs.
3986 Previously, if there was no break point between 40 and 75 characters, Exim
3987 left the rest of the message alone. Two changes have been made: (a) I've
3988 reduced the minimum length to 35 characters; (b) if it can't find a break
3989 point between 35 and 75 characters, it looks ahead and uses the first one
3990 that it finds. This may give the occasional overlong line, but at least the
3991 remaining text gets split now.
3993 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3994 file descriptors might be low, and that setting 1000 would always raise it.
3995 It turns out that in some environments, the limit is already over 1000 and
3996 that lowering it causes trouble. So now Exim takes care not to decrease it.
3998 40. When delivering a message, the value of $return_path is set to $sender_
3999 address at the start of routing (routers may change the value). By an
4000 oversight, this default was not being set up when an address was tested by
4001 -bt or -bv, which affected the outcome if any router or filter referred to
4004 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4005 log only when "message" or "log_message" are set. However, if one of the
4006 conditions was an address verification, or a call to a nested ACL, the
4007 messages generated by the underlying test were being passed through. This
4008 no longer happens. The underlying message is available in $acl_verify_
4009 message for both "message" and "log_message" expansions, so it can be
4010 passed through if needed.
4012 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4013 new expansion $bh_ to give the encoded byte string without charset
4014 translation. Translation happens only if iconv() is available; HAVE_ICONV
4015 indicates this at build time. HEADERS_CHARSET gives the charset to
4016 translate to; headers_charset can change it in the configuration, and
4017 "headers charset" can change it in an individual filter file.
4019 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4020 that creates RFC 2047 encoded "words" labels them as that charset instead
4021 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4022 expansion operator; (ii) when Exim creates a From: line for a local
4023 message; (iii) when a header line is rewritten to include a "phrase" part.
4025 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4026 buggy, causing it to skip the first lines of messages whose message ID
4027 ended in 'D'. This would not have bitten before Exim release 4.14, because
4028 message IDs were unlikely to end in 'D' before then. The effect was to have
4029 incorrect size information for certain domains.
4031 45. #include "config.h" was missing at the start of the crypt16.c module. This
4032 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4035 46. If there was a timeout during a "random" callout check, Exim treated it as
4036 a failure of the random address, and carried on sending RSET and the real
4037 address. If the delay was just some slowness somewhere, the response to the
4038 original RCPT would be taken as a response to RSET and so on, causing
4039 mayhem of various kinds.
4041 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4042 when I implemented it. It didn't allow for the fact that some option values
4043 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4044 the right test for positive values.
4046 48. Domain names in DNS records are case-independent. Exim always looks them up
4047 in lower case. Some resolvers return domain names in exactly the case they
4048 appear in the zone file, that is, they may contain uppercase letters. Not
4049 all resolvers do this - some return always lower case. Exim was treating a
4050 change of case by a resolver as a change of domain, similar to a widening
4051 of a domain abbreviation. This triggered its re-routing code and so it was
4052 trying to route what was effectively the same domain again. This normally
4053 caused routing to fail (because the router wouldn't handle the domain
4054 twice). Now Exim checks for this case specially, and just changes the
4055 casing of the domain that it ultimately uses when it transmits the message
4058 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4061 50. If a filter generated a file delivery with a non-absolute name (possible if
4062 no home directory exists for the router), the forbid_file option was not
4065 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4066 the existing equality matching.
4068 52. Exim was using ints instead of ino_t variables in some places where it was
4069 dealing with inode numbers.
4071 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4072 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4073 called TMPDIR, and if it finds it is different, it changes its value.
4075 54. The smtp_printf() function is now made available to local_scan() so
4076 additional output lines can be written before returning. There is also an
4077 smtp_fflush() function to enable the detection of a dropped connection.
4078 The variables smtp_input and smtp_batched_input are exported to
4081 55. Changed the default runtime configuration: the message "Unknown user"
4082 has been removed from the ACL, and instead placed on the localuser router,
4083 using the cannot_route_message feature. This means that any verification
4084 failures that generate their own messages won't get overridden. Similarly,
4085 the "Unrouteable address" message that was in the ACL for unverifiable
4086 relay addresses has also been removed.
4088 56. Added hosts_avoid_esmtp to the smtp transport.
4090 57. The exicyclog script was not checking for the esoteric option
4091 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4092 will work only if exicyclog is run under the appropriate euid.
4094 58. Following a discussion on the list, the rules by which Exim recognises line
4095 endings on incoming messages have been changed. The -dropcr and drop_cr
4096 options are now no-ops, retained only for backwards compatibility. The
4097 following line terminators are recognized: LF CRLF CR. However, special
4098 processing applies to CR:
4100 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4101 nor a local message in the state where . is a terminator.
4103 (ii) If a bare CR is encountered in a header line, an extra space is added
4104 after the line terminator so as not to end the header. The reasoning
4105 behind this is that bare CRs in header lines are most likely either
4106 to be mistakes, or people trying to play silly games.
4108 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4109 was being incorrectly given as 18 bytes larger than it should have been.
4110 This is a VOB (very old bug).
4112 60. This may never have affected anything current, but just in case it has:
4113 When the local host is found other than at the start of a list of hosts,
4114 the local host, those with the same MX, and any that follow, are discarded.
4115 When the list in question was part of a longer list of hosts, the following
4116 hosts (not currently being processed) were also being discarded. This no
4117 longer happens. I'm not sure if this situation could ever has previously
4120 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4123 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4124 header. This is something that is recommended in a new Internet Draft, and
4125 is something that is documented as being done by Sendmail. There are two
4126 possible values. For messages generated by the autoreply transport, Exim
4129 Auto-Submitted: auto-replied
4131 whereas for all other generated messages (e.g. bounces) it adds
4133 Auto-Submitted: auto-generated
4135 63. The "personal" condition in filters now includes a test for the
4136 Auto-Submitted: header. If it contains the string "auto-" the message it
4137 not considered personal.
4139 64. Added rcpt_include_affixes as a generic transport option.
4141 65. Added queue_only_override (default true).
4143 66. Added the syslog_duplication option.
4145 67. If what should have been the first header line of a message consisted of
4146 a space followed by a colon, Exim was mis-interpreting it as a header line.
4147 It isn't of course - it is syntactically invalid and should therefore be
4148 treated as the start of the message body. The misbehaviour could have
4149 caused a number of strange effects, including loss of data in subsequent
4150 header lines, and spool format errors.
4152 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4153 client host had authenticated. This control can now be exercised by an ACL
4154 for more flexibility.
4156 69. By default, callouts do not happen when testing with -bh. There is now a
4157 variant, -bhc, which does actually run the callout code, including
4158 consulting and updating the callout cache.
4160 70. Added support for saslauthd authentication, courtesy of Alexander
4163 71. If statvfs() failed on the spool or log directories while checking their
4164 size for availability, Exim confusingly gave the error "space shortage".
4165 Furthermore, in debugging mode it crashed with a floating point exception.
4166 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4167 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4168 serious problem, Exim now writes to the main and panic logs when this
4169 happens, with details of the failure. It then refuses to accept the
4170 incoming message, giving the message "spool directory problem" or "log
4171 directory problem" with a 421 code for SMTP messages.
4173 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4174 0, 1, and 2 exist, because some OS complain for execs without them (see
4175 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4176 descriptors. However, the code omitted to check that the open succeeded,
4177 causing mysterious errors if for some reason the permissions on /dev/null
4178 got screwed. Now Exim writes a message to the main and panic logs, and
4179 bombs out if it can't open /dev/null.
4181 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4182 interact so that it is all more flexible. It is supposed to remain
4183 backwards compatible. Also added extra_local_interfaces.
4185 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4186 to bomb out with an assertion failure - to the client this appears as a
4187 connection drop. This problem occurs in the part of the code that was taken
4188 from the Samba project. Fortunately, the assertion is in a very simple
4189 function, so I have fixed this by reproducing the function inline in the
4190 one place where it is called, and arranging for authentication to fail
4191 instead of killing the process with assert().
4193 75. The SPA client code was not working when the server requested OEM rather
4194 than Unicode encoding.
4196 76. Added code to make require_files with a specific uid setting more usable in
4197 the case where statting the file as root fails - usually a non-root-mounted
4198 NFS file system. When this happens and the failure is EACCES, Exim now
4199 forks a subprocess and does the per-uid checking as the relevant uid.
4201 77. Added process_log_path.
4203 78. If log_file_path was not explicitly set, a setting of check_log_space or
4204 check_log_inodes was ignored.
4206 79. If a space check for the spool or log partitions fails, the incident is now
4207 logged. Of course, in the latter case the data may get lost...
4209 80. Added the %p formatting code to string_format() so that it can be used to
4210 print addresses in debug_print(). Adjusted all the address printing in the
4211 debugging in store.c to use %p rather than %d.
4213 81. There was a concern that a line of code in smtp_in.c could overflow a
4214 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4215 initially expressed, the concern was not well-founded, because trailing
4216 spaces are removed early. However, if the trailing spaces were followed by
4217 a NULL, they did not get removed, so the overflow was possible. Two fixes
4220 (a) I re-wrote the offending code in a cleaner fashion.
4221 (b) If an incoming SMTP command contains a NULL character, it is rejected
4224 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4225 runs initgroups(), so that if the Exim user is in any additional groups,
4226 they will be used during message reception.
4232 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob